Nick
@nick@shore.me.uk
157 following, 242 followers
Many have already taken advantage of our early bird pricing for the #MikroTik Professionals Conference but it may be prudent to buy your tickets before the price goes up at the weekend! 😉 We are also very interested in hearing from anyone who wishes to present a MikroTik related topic. https://mtpc.world/ #MTPC #Conference #Prague
MS is having DNS issues. So the Dutch railways are having issues. Critical national infrastructure should not be on cloud systems owned by foreign companies. This is resiliency 101.
@quixoticgeek
actually shouldn't be on Cloud at all.
A National railway can afford its own servers.
@raymaccarthy
Not so sure about that with the Dutch Railways.
@quixoticgeek
@reinouts @raymaccarthy NS should be able to afford it. Whether they choose to, that's a different question.
@quixoticgeek @reinouts @raymaccarthy they can definitely get it at a couple of national hosting companies for a reasonable and predictable price.
@wsslmn
Well I know for a fact that they are heavily invested in Azure. Whether they can afford to switch, if they would want to, is dubious: NS are making structural losses since COVID and have had to make tough choices.
@quixoticgeek @raymaccarthy
@reinouts @wsslmn @raymaccarthy NS shouldn't need to make a profit. The purpose of public transport is to transport the public. It's a choice by the sole shareholder of NS, the Dutch government, to operate this way.
@quixoticgeek
I couldn't agree more. Yet, the reality is that budgets are tight for NS.
@wsslmn @raymaccarthy
@reinouts @wsslmn @raymaccarthy artificially. They don't need to be.
@nick @quixoticgeek @reinouts @raymaccarthy people often underestimate the actual cost of cloud services until something starts to use a bunch of resources, or your setup is so complex that moving away becomes impossible.
@quixoticgeek @reinouts @raymaccarthy
I'm not convinced it's even a case of being able to afford it. It is very likely that the cloud was sold to them as cheaper, but in the end I suspect it may actually end up more expensive due to various auxiliary cost management costs.
However, switching back may now be very difficult (and expensive) as they would have to rehire all the on-prem and Colo capable engineers. As lock-in strategies go, this is not a bad one.
@reinouts @raymaccarthy @quixoticgeek To be clear, running a server is, *relatively speaking* not that expensive. Especially when it's serving a relatively limited area. (It isn't as if they have to have their servers handling traffic across the world.) It can seem a bit prohibitive to us end users, but even for a struggling government a basic server setup isn't *that* big.
That said, anything as fundamental as a railway should be considered fundamental and be required to be self-sufficient.
@reinouts @raymaccarthy @quixoticgeek I ran out of characters, but to be clear, what I mean about being self-sufficient being fundamental is that it's a part of the thing itself and shouldn't be considered separate. That is, the systems running the railways are *part* of the railways and if they can't have the systems it's like not having tracks. You... don't run the rails without tracks.
@quixoticgeek in finance we have DORA legislation for digital resilience (you need to have a backup plan to stay running during such an event), but where are the same requirements for other infrastructure?!
@quixoticgeek @HonkHase whyyyy? 
@quixoticgeek @g0rb @HonkHase Their old stack was a bunch of disconnected services. They noticed and obviously loved PowerBI, and went all-in on connected *every* sensor and data stream they had and created afterwards in their Azure stack. There are some presentations and articles on this from ± 2018 (mid-transition), I can look for these later
@quixoticgeek so uh. As someone trying to do this for critical infrastructure: it actually probably should otherwise it will be down a lot more. At least in the current environment.
We should probably ask ourselves how to get out and still have the uptime, but it would need a far more honest look at our tech and how we fund it.
I cannot see that happening.
@Di4na as someone who's built a career out of building resilient computer systems. It can be done. But you need to have the knowledgeable staff to make it happen. And quite frankly. Those skills have atrophied across the industry.
@quixoticgeek i think they have not atrophied. We never had them that much. Because at no point in the past did we really handle that scale for so many different systems.
This is all new and we never built it.
And I have seen the results of the people telling me it can be done. Their uptime is not better, etc.
@Di4na the reality is that most organisations are not "that scale". So so so many companies massively over complicate everything.
@quixoticgeek wait what? That was what was up? I heard about issues with paying by card at the ticket machines (card still works for the barriers/to tap in and out) but I had no idea it was connected with a Microsoft DNS issue. Missed the first words of the announcement
@quixoticgeek
The Scottish Parliament has had to suspend business.
Literal democratic process is being affected.
@quixoticgeek
Trains were running mostly okay. It was just the App that wasnt working. With NS that's not that big an issue, since most trains run regular schedules. (If that happens to DB and their bizarrely irregular schedules and delays, it would have been a lot more of an issue to still get somewhere...)
@j2 and ticket machines. And ticket purchases. And OV fiets.
At Amsterdam Zuid I saw someone unable to travel cos they were unable to buy a ticket for their bike. Real world tangible effects.
@quixoticgeek Hardly anybody needs ticket machines anymore these days. You can travel with any bankcard. And I'm pretty sure that if you couldnt buy a ticket for your bike they'd still let you on the train.. It's the Netherlands: if the ticketing system doesnt work everybody travels for free.
@j2 I literally watched someone turned away because they couldn't buy a bike ticket.
@quixoticgeek Like on the actual train!? Wow. I think you'd stand a pretty good chance of filing a complaint about that.
@j2 @quixoticgeek a complaint? To get what? With what proof?
The fact stands: critical infrastructure shouldn’t rely on foreign providers. I’d argue that they shouldn’t rely on any kind of Internet connection at all, but the Dutch system as a whole would have big issues implementing that — I’ve seen the temporary OV chipkaart validators for NS buses with their big batteries and 4G cells, it’s a complete nonsense.
@j2 @quixoticgeek (also the part about « everyone travelling for free when it’s not working in the NL » is wrong, I’ve already been prevented from entering the station at Almere Centrum when some local incident made the validators out-of-service for like one hour. Everyone was invited to go away or wait until the issue was resolved.)
@j2 @quixoticgeek Do you know how many tourists we have around here lol
@SIT
Yes. And pretty much any bank card or creditcard will work.
@quixoticgeek
@j2 @quixoticgeek Global adoption of contactless payment is not as high as (Western) Europe, I wouldn't assume it for foreigners
@SIT
Even most Americans have that nowadays. And I assume 99% of tourists are on Google or Apple and could use that if they wanted to. I don't go to Amsterdam that much anymore, but even in The Hague or Utrecht I rarely see tourists at the ticket vending machines anymore...
@quixoticgeek
@quixoticgeek FWIW after working with some energy net providers (both in Germany and Netherlands), I’m fairly certain that their systems are autonomous.
So I suppose there are levels to criticality.
@quixoticgeek my train is also delayed, wondered if it was the same issue for a second but it turned out to be a much more Devon reason of livestock on the line
@quixoticgeek The Microsoft sales person probably promised it would be but it was more profitable not to keep that promise because they'd never check.
@quixoticgeek I predict it will selectively list, authoritatively, all members of a disapproved class as "not citizens", who then get deported without due process.
@quixoticgeek @cstross it has been proven over and over and over again that relying on the big cloud companies for critical infrastructure is a bad idea because it's putting all your eggs in one basket meaning it ends up not just being trains its trains, banking, taxis, supermarkets, hotel check in/front desk and your apartments shiny new smart lock that the landlord insisted on for some reason. Any one of these things going down would be a pain in the ass but all of them at once is a disaster
@quixoticgeek It's not like consorcium of EU hosting provider existed ?
Public money -> Public code
Eu money -> Eu infrastructure.
Seem too simple for our brib... lobyed official.
@quixoticgeek @stux You'd be surprised (or not) at all of the hospital systems impacted by this, too.
@quixoticgeek back in the day, the web hosting company I worked at had DNS servers set up with 3 different providers in 3 different places. In the 20 years the company existed, they never had a dns outage.
To be fair, AWS's DNS servers were up throughout the outage, they were reliably serving what they'd been told to.
It's just that automated systems had decided to remove a bunch of records from the zone...
@quixoticgeek How stupid can you be.
@fenixmaster me or NS?
@quixoticgeek I should have been clearer. The NS, of course. The Dutch government and companies that have been part of it are all lazy and go for convenience and don't think in terms of being independent from American Big Tech. They still think in terms of Windows 95 and stumbled into the cloud that way.
edit: added "stumbled" construction.
New blogpost: Using a SSH config: https://h3artbl33d.nl/blog/using-a-ssh-config
Hi.
About, the example made on the full config section, it seems to me that there is a structural error. Let me explain.
The man 5 ssh_config wrote:
> Since the first obtained value for each parameter is used, more host-specific declarations should be given near the beginning of the file, and general defaults at the end.
(see on the Description section)
Your example shows the general defaults at the start ;)
Ah, good catch - thanks a million. Edited and corrected.
@h3artbl33d
Excellent! :D
I've added a mention of this to the remarks section: https://h3artbl33d.nl/blog/using-a-ssh-config
@h3artbl33d “An SSH config file is compromised…”
ITYM “comprised”. A compromised SSH config file would not be good.
@h3artbl33d I do this. It makes everything easier. Not just on OpenBSD but also in Linux. I should try in Windows as well
Does anyone actually pay more for 5G services in 2025? The value argument used here by the researchers at PolicyTracker feels like it's several years too late.
About a year ago, a client I've worked with for over fifteen years informed me that some of their "less critical" servers would be migrated to $CLOUDPROVIDER. According to them, this provider would guarantee an efficient management panel, "more freedom for their devs", and lower costs. This didn't impact me financially but, on an ethical and personal level, I warned him about the potential problems. Yet they decided to move forward, aided by the arrival of $YOUNGDEV who "has worked with it, it's reliable, and everything works fine". Again, I warned them (where are the backups? A disaster recovery plan? etc.) but they insisted: $CLOUDPROVIDER is efficient and gives us everything.
I studied their plan and immediately understood that their "cost-cutting" strategy wouldn't work: I know their workloads, and the plan they chose was insufficient. Needless to say, a few days later they went down and had to make an "emergency" purchase of the next tier up. The cost? Higher than their previous server infrastructure.
I heard nothing more about these workloads for almost a year but my monitoring tools still were marking them down, from time to time. Then, I get a phone call this afternoon. $YOUNGDEV asks me for support. He doesn't explain, but I immediately understand it's one of those workloads. A serious problem, and they don't have a backup of the database. They don't have a test environment to run diagnostics. The DB is very large, and they don't know what to do. My predictions - not even my worst ones - had come true.
I was running between two appointments. I only remarked that this situation could have been avoided and that it's not something I manage or can manage, but I nonetheless suggested we sync up tomorrow morning. I'm not going to get my hands dirty, but still, $YOUNGDEV is in trouble, and I offered to take a look to suggest a strategy. I then asked for the access credentials to $CLOUDPROVIDER, considering that up until a year ago, I managed all of these workloads. He replied that he "doesn't know if he can give them to me" and that he "would have to ask his bosses". I pointed out that if he wants my help, I need something - I don't even know how $CLOUDPROVIDER grants access to data (or if it does) - how can I give him advice?
It's 18:30 and I have received nothing. Tomorrow morning, if the phone rings, I will answer, but at this point, I won't do anything. I prefer, albeit reluctantly, to completely end the relationship with this client.
If this is the price of dignity and respect, I'll gladly pay it.
@stefano It is the price of integrity. You could always give the client a price sheet for returning the services on prem at their previous level of reliability. They would probably pay for it by not having to pay for $YOUNGDEV any more.
I was always surprised when the biz types were amazed that I knew more about their business and costs than they did. I have a hard time relating to people who just learn surface details and act on those.
@ChuckMcManis What has always surprised me about some companies is that their eagerness to do something "new" or "different" drives them to destroy what made them prosper for years, in pursuit of what is clearly a utopia upon even a basic analysis. But I've gotten used to it by now.
@nick @ChuckMcManis exactly - no backup plan, no disaster recovery - and I'm not talking about data, this time.
@stefano I've found this is often driven by a manager's desire to 'have an impact' which, for them, means "improves the business" not "keep it going." Their analysis on everyone I've dug into looked entirely at the possible benefits and *always* minimized and risks or downsides. For new managers especially "do what the old manager did" is not a recipe for moving up the ladder.
1/2
@stefano this!
Also it seems that a lot of those same people have no grasp of their actual operating costs of on-prem equipment. I have reviewed the IaaS cost sheets for the workloads they want to transfer, and when remarking this is app. 4x what our on-prem cost is the only thing I got is a bewildered stare. Sometimes followed by "but we don't need the people anymore". Sure, VMs in aws or azure administer themselves and backups just happen.
Complete lala land.
@fedops @ChuckMcManis I agree. And the scariest thing is: when they go "to the cloud (TM)", they just stop caring about backups. And when I ask "where's a backup of your data", they're usually puzzled: they think that in the cloud, you don't need backups.
@stefano much like with "ai" a large part of the current crop of IT manglement absolutely does not understand what "cloud" even means. Or proper IT services outside of the ITIL BS for that matter.
To quote my favorite, Admin Zen: "Nobody wants backups, everybody wants restores. To get restores you need backups."
@ChuckMcManis
@stefano @ChuckMcManis the well worn path has no surprises, including no surprising extra income. New and shiny offers the possibility that more money may magically appear from some new as yet unimagined source
@oxyhyxo @ChuckMcManis ...or it can destroy all your business in a second.
The service they're trying to fix is a service they're providing to a client - one of their oldest clients. And not the most patient ones.
@stefano @ChuckMcManis are you sure it’s the eagerness to do something new? Cases I’ve seen it’s almost always the vague(!) idea of „it’ll be cheaper“ or „it’ll be better“, often based on claims or even promises made by a new employee, contractor or consultant. When the people responsible at the company feel, are old to be or actually are „behind“ on what’s new or en vogue in terms of technology they’re particularly vulnerable to such exploits.
@stefano wow. Whatever decision/route that you'll take, it'll be correct, even if ... seems wrong. I know the feeling though.
@stefano But... the moral higher ground, could be helpin' them out. Not for free though.
@stefano this is more or less how I lost the best job I ever had.
@jpaskaruk because of the movement to the cloud?
@stefano Attempted/Intended movement.
https://www.youtube.com/watch?v=cEp910UTrnM
The large render farm I had already deployed on prem continued to be 100% used, but that thing also happened, using AWS's own "experts" to advise on setup. Said advice resulted in absurd expenses from storage, and especially egress fees from datacentre to datacentre. Meanwhile the enthusiasm for The Cloud was unfightable in the higherups. Same story.
When I got wind of this, I offered to set it up so that the cloud render nodes would get the inputs from our on prem file servers, which had plenty of bandwidth to spare, and write the resulting frame to a small, temporary bucket that would be cron'd back down to our servers and removed from the cloud asap. No fucking egress fees. Was not considered.
We also spent considerable on reserving compute capacity, whereas it would have been feasible to atomize frames into tiny chunks that could have been processed on the higher-risk pre-emptable Compute nodes, the size of the frames submitted based on current response times.
That last paragraph would have involved collaboration from people more directly experience in Animation than I, so it never even got to proof of concept, but it was 100% feasible and could have saved our ass.
A lot of hearts were broken when we went under. I miss working in cartoons.
@stefano I’m guessing $YOUNGDEV is a Full-Stack Overflow Developer and nobody told them about database backups.
You should thoroughly wash your hands clean of any involvement there and have an easier life for it.
@stefano following to see outcome.
Historical unrelated anecdote, small render farm, storage/render-engines/edit-suites all connected by multiple fibre cards. We were hitting bandwidth constraints moving data 20 METRES.
The cloud sales people convinced management that storage/render could magically be done off site. The national infrastructure at that time simply had insufficient bandwidth, not a matter of opinion. I left, I think they limped on with a foot in both camps.
@stefano if you sent them an email a year ago, warning about their cloud setup not being adequate: take that email, paste it with timestamps and all into the reply, and send it off as is.
@stefano It's insane how higher ups complete ignore all potential risks when the cloud is involved. Once they hear "potential savings", it's all over.
But I see the same ignorance every day, where businesses prioritize saving pennies while risking tens of thousands in losses.
@stefano In my case, I always recommend that they get a second internet connection, as the clients I work with run their workloads on-prem. However, rather than investing a few hundred euros in a 4G/5G router every ~5-10 years and paying around 10 euros per month for a SIM card, they risk losing thousands in sales every day when they're offline.
I just can't understand how a business owner could justify taking such a huge risk. It's essentially saving pennies on the dollar. Yet when their DSL or cable does go down, they demand immediate support and somehow blame us for their losses. That's basically like living without health and car insurance.
@subnetspider I agree. A decent 4G router (perfect for a backup connection) is less than 100 euros now, and a sim only is less than 10 euros per month here.
When I propose them as a backup connection solution, they often say it's not needed. Until some random road worker will cut the fiber and they'll stay dark for days.
@stefano I am on the other side of something like this rn. IT middle mgmt in a architecture role and up top direction is cloud-first / cloud-only. Really hard to help steer a ship in such a constellation, especially with the somewhat dooming financial downturn from trade uncertainties.
I think “the cloud” has its value for highly elastic / scaling workloads. Stable workloads should be at least hybrid if not fully onPrem in some Colo - Datacenter operations is not some secret magic.
@stefano I'm getting ready for Halloween, but don't let them do it. Everything is going to the cloud (where we lose control of our own stuff) and AI (so they can learn how our mind motivations and impulses work-for some reason). I will get more into it later, but wantedto make a place mark before I get Hallowed Eved tonight.
New version! #MikroTik RouterOS 7.20.2 is available in the Stable channel https://mikrotik.com/download
So it turned out that the place Sandra booked for a week's "let's have some sun before winter" was, in fact, Magaluf.
I certainly never expected to go to Magaluf.
Of course, I did what all Brits do in Magaluf: photograph Internet and telecoms infrastructure.
Alt...
Photo of a shop sign saying "Public call office" and "Telephone card"
Alt...
Photo of a series of external wall-mounted boxes, providing fibre and cable distribution.
It was fascinating, in an odd way, to see what all the fuss was about.
Out of season, of course - so, so much of the "strip" was closed.
@neil I stayed in the next resort along from Magaluf, in a rainy February in maybe 2004. It was not what I would call "happening".
@neil Left to right: Bo customers, No customers: ALL the customers, No customers, old copper infra.
Weird that none of the 3 "alternative networks" have gained any customers.
@neil I went to Magaluf once. Had a nice sandwich (filling not remembered), walked about a bit and then left. Was under 12 years old at the time. Hope this was a helpful review.
@neil I had no idea. For the rest of us, i.e., non-British...
![screenshot from https://en.wikipedia.org/wiki/Magaluf#Nightlife
"The nightlife and associated activities results in the town being commonly nicknamed "Shagaluf" and “Megamuff” in popular British culture.
The resort focuses almost exclusively on British binge drinking tourism, with extremely low alcohol and fast food prices, and British staffed medical emergency centres located close to bar areas.[2]
British tour operators have warned the Spanish Tourist Board that the image of Magaluf[8] is affecting their efforts to market Mallorca"](https://media.mas.to/media_attachments/files/115/447/847/072/417/406/original/94485ac5146be717.png "screenshot from https://en.wikipedia.org/wiki/Magaluf#Nightlife
\"The nightlife and associated activities results in the town being commonly nicknamed \"Shagaluf\" and “Megamuff” in popular British culture.
The resort focuses almost exclusively on British binge drinking tourism, with extremely low alcohol and fast food prices, and British staffed medical emergency centres located close to bar areas.[2]
British tour operators have warned the Spanish Tourist Board that the image of Magaluf[8] is affecting their efforts to market Mallorca\"")
Alt...
screenshot from https://en.wikipedia.org/wiki/Magaluf#Nightlife
"The nightlife and associated activities results in the town being commonly nicknamed "Shagaluf" and “Megamuff” in popular British culture.
The resort focuses almost exclusively on British binge drinking tourism, with extremely low alcohol and fast food prices, and British staffed medical emergency centres located close to bar areas.[2]
British tour operators have warned the Spanish Tourist Board that the image of Magaluf[8] is affecting their efforts to market Mallorca"
@neil Are you going to to it the authentic way by downing 10 Stellas in the airport, having a punch up and get denied boarding?
Implemented more scopes to match other ActivityPub implementations (public, unlisted, followers-only and direct message) (contributed by byte).
New icons showing instance and actor failures.
Mastodon API: Added remote accounts follow metrics and statuses when viewing profiles (contributed by Stefano Marinelli), fixed post deletion.
Fixed outbox collection (contributed by byte).
New file FEDERATION.md (contributed by andypiper).
Updated Czech, Spanish and Brazilian Portuguese translations (contributed by pmjv and daltux).
Fixed manpage typos (contributed by r-ricci).
If you find #snac useful, please consider buying grunfink a coffee or contributing via LiberaPay.
And thanks to byte, @stefano@bsd.cafe, andypiper, pmjv, daltux, r-ricci and any others I may have missed for their contributions to this release as well!
I submitted a Pull Request to update MacPorts' snac to 2.84 here:
https://github.com/macports/macports-ports/pull/29810
1 out of 3 of GitHub's Continuous Integration checks passed, which is a good sign the other two will pass as well.
It's up to someone else with commit access to merge it.
#snac #MacPorts #OpenSource #ActivityPub #Mastodon #NoDatabaseNeeded
#NoJavaScript #NoCookiesEither #NotMuchBullShit
I've a feeling that I will need to make a couple of quite big changes to my computing in the near(ish) future, from the perspective of more people asking uncritically for Cyber Essentials as a base line:
* Debian release: I will likely move from Debian testing back to Debian stable, because of security patching policy/schedule. Or perhaps sid, but that feels harder to explain. Moving back to stable will be a pain, as there is no downgrade path.
* Hardware: from cheap old Intel NUCs as servers, which suit my purposes just fine but which no longer get firmware updates, to machines which still get firmware updates. I'm not massively convinced of the threat posed by this, but hey ho.
For the Debian release, I could move all machines using "testing" to "forky", to track the forky release and then, when forky becomes stable, I'm sorted. But that is - AFAIK - a 2027 issue, which might be too long to wait.
Firmware: looking at the detailed requirements, it looks like this relates solely to "firewall and router firmware".
So the NUCs would be fine.
I'm probably screwed in terms of the requirements that software must have a "vendor", and that that vendor:
> has committed to support by providing regular vulnerability fixes. The vendor must provide the future date when they will stop providing these.
Not compatible with most FOSS.
@neil Is this one of those cases where you can document the shortcomings of and why you don't support CE or is your hand forced here?
@dentangle I will have to see. I started on this a while back, and got distracted, and now there are updates requirements.
I will do my best to be appropriately secure, and I hope that I am *much* better than many organisations, but compliance with specific standards may be challenging.
The 'Vendor' angle is interesting. Does that suggest only using a distribution like RH or Ubuntu etc with a major org behind it, compared to, say, Debian. Most definitions of "vendor" include a reference to selling - although the above 'vendors' don't sell the OS. Ah, well, I'll let you ponder. What a kerfuffle.
@dentangle @neil CE assessors are notoriously pedantic. You get 3 goes at resubmissions if they're unhappy; after that you pay (and start) again.
For example; they won't accept the inbuilt MacOS virus scanner as Apple don't publish the update cycle for the definition database.
@hugh @dentangle At which point, it feels (even more) performative rather than actually useful...
@neil @hugh @dentangle which is why we went to 27001, as at least this allows you to define your risks. Not sure how that scales in terms of costs to a business like yours though. The documentation would be a pain!
@glaringanomaly @hugh @dentangle
I'm working through 27001, and I have written a lot of documentation.
But 27001 really doesn't deal well with such a small business.
@glaringanomaly @hugh @dentangle Plus, as Hugh says, there is a push for CE. I'd like to be able to tick that box if I can.
@neil If you use Ubuntu then you have a vendor for your Debian software who has a published support period. On the firmware though, we have a requirement that all computers must me recycled after 5 years as they won't have firmware updates.
@DanielRThomas Debian here :)
@neil @DanielRThomas there must be 3rd parties you can pay for a support contract for Debian… would that cover it?
@pauldoo @DanielRThomas It might, but at that point it is just another pointless compliance payment, for the pleasure of doing business. So I would rather avoid that route if I can.
@neil @DanielRThomas sure, probably pointless. Is there a qualification or accreditation you could get that would allow you to claim that you yourself are the support vendor? :D
Now, funny that you too should be thinking along those lines (although, from an ethics point of view, I would not go down that route).
@neil @DanielRThomas yep. Whether your vendor is Microsoft, Apple, Canonical, or a 3rd party, it doesn’t at all guarantee that security issues will be fixed promptly. None of those companies even wrote all the code they vend to you and claim to support. Having an official vendor provide support is about you being able to show you did make a professional best effort to have the risks covered, and weren’t trying to yolo it yourself ?
@neil @pauldoo @DanielRThomas it might not be that pointless if they were actually contributing (via labour or money) to making your software more secure. I wonder if there is a business model there for a lot of small businesses or coops
@neil I suspect this is why a lot of places go for RedHat, Ubuntu or similar: the (often pricey) support contracts.
Would you imagine a third party support contract (so not with Debian) combined with Debian release cycle docs would be enough?
Or because it is "third party" it automatically doesn't count as the vendor?
@tyrthecat I will need to dig into it, tbh.
I can easily find an organisation which will make a contractual commitment to do whatever CE requires. The fact that, in practice, they will not, is what makes it problematic.
I have Cyber Essentials certification for my Linux machines (Ubuntu, Pop-OS, Fedora).
The former two seem to be "vendors", since they are companies. The latter may be close enough to RedHat to qualify.
@neil
Debian does publish EOL dates for security updates, and does push updates until then. That's not enough to satisfy the "vendor" requirement?
@neil In my last job, we had CE (and CE+, and 27001), and yet we developed software for UK government use using almost entirely FOSS tooling. Everything was deployed on Ubuntu in Azure, and several developers used Ubuntu and/or Debian within WSL.
So it's definitely possible. I don't have contact details for the compliance person any more (and the company's gone now, too(*)), so I can't ask for the details, I'm afraid.
(*) For reasons unrelated to security standards compliance.
@neil IME of helping to assure a 2500 person enterprise to CE+, our auditor was fine with our Linux distros so long as they were receiving updates, and we could prove we were applying them in a timely fashion.
I don’t recall a ‘vendor’ requirement. That may have changed in the last 2 years though?
That is really positive. Thank you.
> I don’t recall a ‘vendor’ requirement. That may have changed in the last 2 years though?
"vendor" has been in there for quite a while, but I am hopeful that there is a pragmatic approach to its interpretation.
@neil We were fortunate that our auditor worked with us to interpret the requirements. We found there was a good deal of pragmatism available - though also some silliness (like having to pen-test a sample number of machines running every OS version & patch state on our estate!)
@neil No non-trivial software house could pass CE+ if the standard didn’t accommodate a wide range of Linux OSs - so in practice it does. Otherwise the government could buy no software consultancy!
Let's take Uptime Kuma.
I use it to monitor service downtime. It alerts me to outages, so that I can intervene if needed.
It is an easy to use, clever bit of FOSS.
Is there a guarantee of updates? No! It is FOSS. No warranty etc.
But to *not* use it seems less wise than using it and accepting it for what it is.
Could I bring the machine out of scope? It is not addressable *from* the Internet. But it connects *to* the Internet to get updates. I guess I could consider a proxy for all machines with similar requirements...
@nick Yep, I have egress rules (and ingress rules), and everything is VLAN'd according to what it does. But that doesn't solve the problem here.
@neil @nick I can't speak entirely accurately here as I didn't do the arguing with the assessors, but at work we had the argument accepted that FOSS packages which had active maintenance and were actively updated in response to reported security issues had a "vendor". This covered both the OS (at the time CentOS before RH/IBM blew it up as an Enterprise option, now Oracle Linux) and external packages or locally compiled FOSS packages like Exim.
We also have a CE+ network but that's separate.
@neil @nick For updates, we apply package updates every two weeks or immediately (after assessment of risk) for critical/security updates. That also satisfied the assessors.
I understand that the people doing the assessments can have different opinions but pragmatism seemed to be very much the order of the day for us.
@neil I'm pretty certain that Red Hat or similar would count as a vendor; you're buying the software and support from them. I also wouldn't get too hung up on "firmware". A firewall on a PC running RHL would likely count as "firmware".
What they really want is evidence that any holes in your security get patched promptly.
The standard probably has a definitions section. Read those definitions carefully. They may well have a broader meaning than a first glance would suggest.
@neil I'm literately wrapping up a CE (not CE+) assessment for my company right now and FOSS is explicitly marked as OK at some point. I'll check the wording and which question it's mentioned at and will let you know.
@neil I would *assume* that you can mitigate any potential issues through the use of other controls, depending on quite what the Forky machines are exposed to?
@SonOfSunTzu AIUI, "compensatory controls" are not a thing with Cyber Essentials.
@neil true, but I think assessors are expected to, well, *assess* the context of what's used. It would certainly be good for you to get an opinion from them on your use of Forky before you start changing the version you use. In my humble and uncontracted opinion anyway.
@SonOfSunTzu I've heard varying experiences, and so I think that you are right, that it comes down to the assessor. Some have found it unduly prescriptive and uncompromising, while others have said that they found it a pragmatic and sensible, even straightforward.
To my mind - and I might be biased - I think that we do pretty darned well in terms of security...
@neil I would be interested to know how many businesses are routinely installing firmware updates on their hardware server hardware let alone their end user devices. I suspect very few.
@neil What sort of vulnerabilities could arrive from using out of date firmware? (My server is also hosted on a old nuc)
@neil@mastodon.neilzone.co.uk You'll end up with a never ending cycle of replacing perfectly good hardware every 2 years just because they stopped releasing 'BIOS' updates.
@petererer Frustratingly so.
@neil @petererer which, of course, isn't something either the Cyber Essentials scheme or UK Gov would want, and I think that's worth keeping in mind as you navigate this.
@neil
I've got myself into a bit of a logic hole here re: hardware.
Do the requirements mean you are forced to run hardware that you KNOW must have vulnerabilities (because you also KNOW that something will be patched by a future firmware update?) You won't know what they are at the time of purchase of course, so can't mitigate for them in other ways.
@neil you'll be fine. I help companies get through cyber essentials all the time, generally more obscure setups get through easier amusingly.
> I help companies get through cyber essentials all the time
On a consultancy basis, by any chance?
@neil you could of course do what I see all the time, and reduce the scope so it's completely pointless, and a tick box excercise
*runs*
@neil Last time I renewed my very-small-business (1 person!) Cyber Essentials, I had a lengthy to-and-fro with the assessor about who would "support" my OpenWrt install.
I thought this was about the software, and wrote up a lengthy explanation about my maintenance approach. But it turned out they were obsessed with the hardware. I tried to explain that my router was just an inexpensive generic box with readily available replacements, and that in fact I already had a spare. Not good enough! Who is your maintenance contract with? What is the warranty?
Already looking forward to explaining all about self-hosting NextCloud when the next renewal comes around. Will it make any difference that I am professionally advising people on "cybersecurity", I wonder? :-)
@m Yes, these are exactly the kind of things which I'm looking at now. None of which have any material bearing on our cyber security.
I just discovered how good MikroTik routers (all devices in general) are.
RouterOS 7 looks amazing and the web UI is clean, powerful, and fast. Can’t believe I wasted money on cheap TP-Link and Cudy routers that broke when I tried to install OpenWRT. MikroTik gives you real control, pro-level features, and solid firmware right out of the box.
Wish I knew about them earlier. 🥹
@nick Already using it. They also have a mobile app called Mikrotik Pro. It has the same features as the winbox app!
I'm pleased as the Non Executive Director Candidate I voted for at Nominet was elected.
Nominet is responsible for the .uk registry. That's webdomains.
So when you buy your UK domains from your chosen registrar, it's a Nominet Member you are buying it from.
While it's mostly Hosting Companies and Domain Managers who are members, anyone can be a member.
At the moment Nominet has around 2100 members and only 192 of those members voted. So we have a voter apathy issue.
It is pricey to be a member. But as well as discounted domains, you do get to have a say as a member.
So perhaps organisations like Hackerspaces and Tech coops could consider becoming a member. Especially if some of you have a mild domain addiction and many of them are .co.uk
Part of the reason members are down I think is related to smaller hosters being bought by larger hosters.
We're back at the IPv4 problem again as you can't easily set up a new hosting company without IP ranges
But even though I am a smaller member, I still vote in the elections for the organisations I am a member of.
We have a microcosm here of the larger picture. Members don't have to vote, so many don't.
But the larger registrars do. Much like those for the status quo, or to disrupt the status quo further to the right, they know they need to just keep voting.
In the UK in the last General Election just over fifty percent of the population voted. Some aren't eligible, but there are many who are don't vote.
Which is the danger zone, where Reform can get in.
So when you're annoyed with folks like Green Party voters saying they will vote green and not the status quo. Consider how you make the case to the folks who don't vote.
Find out why and see if you can persuade them to vote, for anyone.
Ask questions, don't demand, help them register and vote. If they have blockers to being able to vote. Consider how to help them. Like drive them to the voting booth. Help them to register.
It's good community practice.
@onepict I was going to stand as a Nominet NED a while ago, but was rather put off the idea.
@neil Nominet has had issues for years. I can see some improvements and I think a lot of that was Kieran McCarthy and Simon's work.
But the fact it had to come to the members leading a campaign to nudge it isn't great.
It is pricey to become and maintain the membership as well. But I think it would be a good thing if more folks outside the traditional hosting and domainer markets could join.
@neil From the memory of that time , it was interesting to see how the board at that time didn't understand how public relations on twitter worked with it's community.
Which is something I've seen repeated on the internet since.
@onepict Yes, I was looking at the time of Simon's campaign.
@neil I can so see why you'd be put off.
I was put off so much by the board at that point in time.
@onepict Such a shame.
@neil I've never considered sitting on a board like Nominets mainly because of the lack of experience.
But I suspect there's a lot of members who feel the same, even when they arguably to have relevant experience.
But then that's an issue faced elsewhere by other boards.
@neil I mean I also don't have the time at the moment. But that's another issue 🤣
@onepict The power given to the big domain holders is the problem there. They have excessive power and therefore control, so those with only a few votes (comparatively) just don't bother.
It's been this way for years, and it'll never be changed as long as the big companies have their way. And as consolidation happens, the problem gets even worse.
(For completeness: I used to work for the Brealey brothers at Webfusion, a long long time ago).
@greem Yeah there's a lot of power with the voting especially as the votes each org gets is tied to the number of domains that member has with nominet.
Which is another reason for folks to look beyond the likes of 123 reg and GoDaddy.
@greem with it costing 400 quid to join and 100 quid per year (both ex VAT) it is pricey.
You do get a discount on UK domains and being able to have a say in one of the main pillars of our digital infrastructure is a good thing.
A limited say, but given the voter apathy with existing members, we could do with a more diverse set of members. After all this org does set base pricing for the main UK domains. There's more of a chance of a say in things compared to say .com or other TLDs.
@onepict Ah, 123-Reg, the brand I can never unsee.
Implemented back in 2001/2 (I think!) by a colleague on the programming side and me on the sysadmin side, under the aforementioned's stewardship.
We pushed a lot of boundaries with that, in terms of DNS. I found the limits of filesystem IOPS, ISC Bind's memory handling and a few other things when it got to a high volume. Restarts took *forever*. Ended up building a load balanced MySQL backed system which could cold start in less than a minute!
“People are the most annoying people.”
-Kemi @shitkemisays
Lots of news printing the story about the restaurant owner getting upset that nobody would eat there.
Not one of them posted its hygeine rating.. https://ratings.food.gov.uk/business/1245963/don-ciccio---osteria-siciliana
I didn't even know the score went that low.
That will be why nobody went..
@tony They really can get that low - there was a curry house not all that far away from me who achieved this prestigious score once.
What does need changing is that in England it's not compulsory to display the rating in the premises itself for everyone to see.
The grid is bursting with green power this weekend, and you know what that means…Free Electricity Session Move any heavy power activity into that hour; be it laundry, tumbledryer, extra electric-heat. The power you'd typically use costs the same. If you’re an Read the
Friday 24th October, 9pm - 10pm UK
AND
Saturday 25th October, 12 (noon) - 3pm UK
The hastily crafted (without checking with the product team first 😇) GreyNoise IP checker tool — https://check.labs.greynoise.io/ — was received pretty well at work today, so I spiffied the CSS up a bit and gave it proper opengraph tags and dark mode support.
Shld be a friendly way for anyone, including non-cyber folks, to see if their home IP or work NAT IP was found in GreyNoise.
Might be a good idea to visit it when you're doing on-site family/friend IT support during the upcoming holiday season.
@Geri 🤞
@sodslawyer I watched #BbcQuestionTime last night. Mad Nads was representing #Reform. The audience was laughing at her.
It's that time of the year again, the National Trust AGM where council members are selected, and yet again, the right leaning Restore Trust have a selection of plausible candidates entered. Be warned, they do not have the best interests of the Trust in mind, more of a culture war. So again, I appeal to all Trust members in England and Wales to make sure you vote. Just go to the site and search for AGM. Enter your membership details. Safest bet is to use the block vote. Please boost.
The client has a terrible, unreliable FTTC connection. So, this morning, I've been testing several devices and 4G carriers to find a reliable alternative when the FTTC is full/down.
The good, old LTE12 Chateau is the best for this task, giving a stable and reliable 220/50 Mbit/sec thanks to carrier aggregation.
The newer but smaller hAP ax lite LTE6 is still giving a good result, around 100/50 - expected, as it's "only" a LTE6. This will probably come to my office, while I'm waiting for the Chateau 5G (LTE20) to arrive - probably not before middle of November.
Now, I need to create proper queues and rules to manage the two connections - at the moment, I've implemented only the failover and some simple queues on the FTTC.
Lunch time.
#Mikrotik #IT #SysAdmin #MorningFun
Alt...
The MikroTik Chateau, the Mikrotik hAP ax lite LTE6, a 4G usb key and a usbc 2.5 gbit/sec ethernet adapter on a wooden table
@stefano Exactly which WWAN module does the Chateau use?
@apicultor this is the chipset of mine (but it's an old device): EG12-EA
@stefano Oh no, Quectel. Their firmware update process is trash and unreliable (if you can even get the update out of them by begging on their forum) and when their updater says success and leaves your modem bricked (exposing only a QDSS DATA device and nothing else) they tell you to buy a new one.
Go on, ask me how I know.
@apicultor Oh, that's a problem. I've had this for 5 years and upgraded it (it's been serving me at office) and it always worked but...yes, I guess it's not that nice.
@stefano Upgraded the modem firmware, or the Chateau? The Chateau probably does not ever attempt to push a firmware update to the modem.
@apicultor both of them
@stefano How did you update the Quectel?
@stefano i have 2 of those hAP ax LTE6 devices and they are surprisingly good. In my case speed isn't much of an issue, but failover configuration and flexibility were. WiFi was a bonus and it works fine. Yeah, I'm a fan. Connecting remote locations through wireguard and I feel confident I can administer these remotely anywhere.
Thinking about it, I do quite like the terminal.
Not quite ready to abandon the gui entirely, but for git, using ssh to access remote systems or plain distraction free writing in base Vim there's nothing like it.
I prefer simplicity because I want the spare brain cycles for other stuff.
My computer should be under my control, not another source of anxiety or pain.
@onepict I've got a great little setup going to do fiction/picture book authoring in neovim (with pandoc templates for the manuscripts)
I figured that if I use this tool all day for work - why not use it? I've got my custom word count scripts, can use version control properly, can comment, can hide formatting away from my brain that loves to play with it - what's not to love?
@onepict totally get you but terminal still such an alien environment for my brain. Maybe a computer solely running some music software would be the way to go.
@inpc I think it does depend on what you're using it for.
Like I started on the command line, then windows 3.1 and personally I think after excel 5 the computing revolution went downhill.
But then I used to use WordPerfect as a secretarial temp. So even word was a huge annoyance.
But then I am a grumpy old crone yelling at the cloud. 🤣
@onepict I always think the terminal gets the power relationship between person and computer right: you tell it what to do, it does it or tells you why it can't. You may need to formulate what you want, but there isn't that game of hide-and-seek, where did they put that setting, does it still exist, can you change it that you get with GUI.
I dunno, I feel like "Computer says no" is more of a GUI thing.
@toxy @djm62 yeah I remember in secondary school we learned to type on Amstrad processeurs with orange on black screens and those weird disks that weren't five and a quarter or three point five inch floppies.
I always preferred the green on black.
I remember being really annoyed at uni with the Unix systems terminals being configured in light mode.
@djm62 @onepict I think that has more to do with GUIs being the common mode of interaction for mass-market products, rather than being some innate property of them (or conversely, of terminals). Terminal users are just less likely to tolerate such patterns, and less likely to be catered to by those putting in dark patterns etc.
@joepie91 @djm62 yeah like computers for a long time were accessible to very few folks.
Access to computers when I started with terminals, you needed to have cash or easy access to one.
Whereas with the mass market the interface is what you're given.
Having a device you can control and replace parts with is still an expensive thing.
@onepict for git, definitely - when I'm training junior people, I always advise them to learn git on the command line.
Even though git has a fairly poor CLI experience due to inconsistency between verbs, all of the UIs for it have to abstract away *something*, and if you don't understand where that abstraction is, you can run into trouble easily
@cait I got taught the joy of git send email the other day and patching.
So much less hassle. Especially if you're on the same LAN.
In the process of renewing some expired network certifications and I’m struck by how much time is spent on QoS - something I have never implemented or seen implemented.
Ok I tell a small lie, I’ve done it on Wi-Fi where the controller can mark certain incoming traffic as high priority and that can help somewhat if the wlan is busy but end to end QoS? Never touched it, never needed it. I view it as incredibly niche. Am I wrong?
@nick exactly. When does the link get full? Every network I’ve touched has at least 20Gb of uplink capacity from access switches. Campus network capacity is cheap, fancy optics don’t cost anywhere near what they used to. I can see this being an issue with some WAN links but then I’d expect to manage that at the router not the switches. 90% of all traffic these days is internet bound. No end to end QoS there.
@nick I recall a saturated 1Gb ospf link in a university campus network. Various QoS configs had been played with and this had helped the terrible VoIP experience but then we just taped another interface onto it as a LAG and problem solved.
@Wifiwits My experience too. Might be a shiny feature but if the network is being asked to carry more traffic than it can handle it's time IME to upgrade the network, not to sort it into first-class and second-class.
@Wifiwits had it on a (relatively) low bandwith private network, upon which the business was implementing IP telephony. QoS needed to be bulletproof or it regularly fell to bits.
I was quite obsessed with QoS so it fell to me to ensure it was right, and it was, and worked wonderfully.
Policy was mega simple (if it's voice it goes first) but we ensured that was true at every hop in the network including e.g. a switch handing off to a bit of hardware, right down to what the hardware could understand (DSCP, CoS) and what hardware queues it had.
Also once upon a time, did a couple of offsite backup solutions that trickled the traffic upload offsite down to 10kbps if it needed to, so others could use the link. That also worked beautifully, you wouldn't know the backup was smashing the link 24/7.
What does baffle me is these policies (particularly in cisco siwtches that do mountains of policing and multiple queues. Usually so complex that they aren't implemented correctly at all, cos nobody can understand it 😂
Struggling with vibe coding? You just need to get better at prompting. Here is a six video series "course" to teach you how to write the most effective AI prompts and get the most out of vibe coding!
(Do you ever wonder what it would be like if there were some special language for giving commands to computers... a kind of prompting language so they would do *exactly* what you wanted? Maybe someone should invent that.)
@futurebird After reading this post, I've decided to flesh out the ideas a little. And I think I may have conceptualised the future of vibe coding!
First, the problem of inconsistencies in how AIs respond to prompts.
We need better predictability!
So let's create a standard syntax for prompts, so when you enter a given prompt in a particular way, the machine will always respond in a standard way.
We'll call these next-gen prompts "commands".
As long as you know this simple "language" of commands, the machine will always do what you want!
Now, on to the problem of needing complex prompts for complex vibe coding jobs.
Let's take away the guesswork.
Instead of trying to precisely enter one complicated long prompt, let's instead have a series of simpler commands that get executed one after the other.
A "program" of commands, if you will.
So you don't have to re-enter your program on prompts, let's invent a physical media for storing and loading these programs.
Now, the next big elephant in the AI room is resource use.
Let's solve this issue so that the command interpreter "AI" software runs with very few resources. I know this is a stretch goal, but perhaps even an 8-bit microcomputer with just 64K of RAM?
And to really make this catch on, let's make this AI vibe command programming language so simple, even an 8-year-old can understand it.
A "basic" programming language, if you will.
Now I know this "Basic" programming language AI vibe coding system sounds extraordinary.
But I promise you that for just $10 billion in VC funding and total indemnity against copyright infringement, I will happily deliver a 64K 8-bit microcomputer with the BASIC programming language preinstalled!
This is the future of AI folks! Feel the singularity approaching!
Help please.
I am trying to see how far my posts on this platform go.
If you see this post can you please like it.
Please do NOT boost it, as I want to see its natural coverage.
I am thinking of posting more often on here and I am just wanting to see what the current reach is like.
@davidallengreen
Are you sure you don't want boosts, On Mastodon / Activity pub there is no algorithm to promote posts, the only way for a post to spread is by users who follow you, boosting it
@BrianSmith950 I am sure - else I would not have typed it.
I tend not to type things about which I am not sure, unless I say "I am not sure".
@davidallengreen i'm not entirely certain how likes federate. it may be a poor measure.
@davidallengreen Interesting. If I do "Expand this page" I see 7 likes. If I do "Open original page" I see 138 likes (and one boost, not me!). I do tend to forget how compartmentalised the like/fave number is; I think we mostly just see the number on our own server!
@davidallengreen in my head the options are boost or favourite so "like" confused me for a split second but I managed to figure it out :)
Hardly any activity on my instance :(
Alt...
Screenshot to capture number of likes on a post whose author I follow but who is not on the same instance as me.
@davidallengreen Reading from Japan
@EdtheChem Hello!
@davidallengreen Hello! I followed you on the old bluebird place and you were in my pocket when I moved out of the UK and across to a new network. Thanks for the fascinating legal insights!
Loving the Mastodon-style "let me explain" responses.
@davidallengreen Hi. How do you (specifically you) measure reach? (Whatever location the instance I'm on might report I'm in northern England).
@bazbt3 By seeing how many Likes hey back to me. I think that is obvious from my post, and I don’t think it needs to be over-complicated.
@davidallengreen your experiment contains the assumption that "likes" propagate back. I'm not sure how reliably that holds
Thank you for that perfect Mastodon "let me explain" response.
@davidallengreen I'm not explaining. I'm just making sure you know all the assumptions of the experiment
@davidallengreen Greetings from Antwerp, Belgium. Where the old courthouse, still in use, has a "War Justice" entrance.
@davidallengreen “Please do NOT boost it” <at least 2 people boost it>
Wonder if the boosters were bots?
boostedHow big of a fuck-up is going to need to take place for people to realise that putting lots of things all in one place and nebulously labelling it as "the cloud" isn't the best way to architect an "inherently reliable" Internet, from an end-user's perspective.
I don't even begin to see the tide turning.
It didn't turn at any of the fuckups that happened at fastly, cloudflare, etc had.
It hasn't turned with any of the amazon ones.
There have been others.
1/3
boosted2/3
At no point do I see any movement towards "is this such a good idea? Should we, I don't know, mad idea, host our own shit?"
Because every time, the forgetfulness happens rapidly thereafter.
"Oh it's working again!"
"Well, let's not forget the outage!"
"What outage? It's working again!"
3/3
What I have seen is a possibly increasing number of home users no longer wishing to rely on the cloud; people bringing back in house their NAS, or their home automation. But this is niche/hobbyist folks mainly. I really don't feel the same is true of businesses, generally.
@bloor Leaving quite a few people around the world unable to do any work so far today. https://www.bbc.co.uk/news/live/c5y8k7k6v1rt
@TimWardCam @bloor my opinion cloud is good for bursting or short lived workloads.
It still doesn't make sense for stuff running 24/7 at least not at hyper scaler pricing
@TimWardCam @bloor I can see the appeal of stuff like managed databases through if you don't have th expertise in house.
You will pay for the privilege of db as a service though
@bloor Then there are the "local systems" which won't allow you to do anything until after they have successfully checked for updates..Oh, the update server is AWOL? Good luck turning your lights on!
@rbairwell @bloor Several things I have owned have been bricked by the app disappearing from the app stores too.
BT Whole Home disks (the original white ones). lolll, update server is gone and the app won't talk to out of date discs nor will the discs update so are now unconfigurable.
Fancy car DAB / hybrid internet radio I had from RadioPlayerUK (obscure, I know).. lol can't control it and it's stuck on one radio station now because the app no longer exists.
@bloor This is certainly something I'm seriously considering.
I'm not quite on the geek/hobbyist scale, but it doesn't scare me either
You're not wrong, but the calculus for most companies comes down to:
- hosting our own shit is kinda expensive
- requires expertise we don't have (and which would be hella expensive).
Therefore:
- the occasional AWS outage is low risk (they're not actually that frequent).
- and (as far as we can tell) doesn't have that big an impact on our bottom line.
- round of golf, and then drinks after...?
@rogerlipscombe @bloor I'm not at all convinced that the cloud is less expensive. I think there is the /feeling/ that on it's face it seems less expensive, like all outsourcing, but I don't know how many people actually honestly look at the cost and value after the fact ("it's too late now").
Used "wrongly" - which many do - the cloud will be more expensive and less reliable than a decent managed virtual machine or even dedicated server on a network that is not designed to be disposable.
@nick @rogerlipscombe @bloor we love Proxmox and supply and manage VMs for a number of customers on it, at least one of whom spends much more - by an order of magnitude at least - with Azure for markedly worse results.
Cloud trades heavily on turn-key-convenience but boy do they charge for that luxury.
@bloor I think a lot of the businesses are happy to deploy the someone else's problem field to the issue. Blaming Amazon for the outage and accepting this. I've seen this type of outage be a vehicle to get discounts on hosting costs going forward as well through sales channels.
@bloor my experience is if you tell modern software developers things like that they look at you strangely and shoo you into the corner... "OK, grandad, go play with your vintage computing or something whist we Run Fast And Break Things like proper software engineers".
Also: "But then we would have to pay someone to do that! That's so dumb."
looks at cloud computing bill that is higher than an OK salary for infrastructure I could manage with probably under 20% of my time (Not that I actually want to do that with my time TBF, but you could hire a whole competent full time person for less!)
Of course the killer argument is then: "but what if we need to scale to Facebook size overnight!?" 🤷
Cannot win.
@bloor 1/x
There's an element of 'no one gets fired for buying IBM' in this - AWS is a 'safe' default that most customers don't need you to justify using.
The managed services (the CDN, DNS, database, queues etc) are honestly quite solid and need a lot less in-house expertise than running your own.
Cost *is* more than some racks in a DC + hardware, but to do that you need someone to spec it, someone to install it, someone to look after it if drives die in the middle of the night, etc...
@bloor 2/2
It also means you don't need to plan capacity in advance or be trapped trying to run a service on limited resources until you can get hardware - and getting hardware isn't as predictable as we'd like these days either.
There *are* advantages to cloud, but I do agree that companies running their own hosting is overlooked these days.
@bloor I think more of a focus lately is how the Americans have proven conclusively - at great speed, no less - that they can no longer be relied upon, and so there is more interest - particularly among government bodies - to repatriate data away from US interests.
Part of that will hopefully include a reckoning about whether or not 'cheapest' is a wise purchasing decision, but I doubt it.
@bloor for a lot of companies their uptime on managed clouds is vastly higher than they could achieve locally. It also appeals to managers because if it breaks it's someone else's fault
@etchedpixels @bloor > for a lot of companies their uptime on managed clouds is vastly higher than they could achieve locally
That's a good point that I think is often missed in the criticism that follows incidents like this. Everything goes down for an hour or two at some point. If all the services affected today had gone down for twice as long but one at a time on different days, nobody would be talking about it.
A lot of skills for IT management/system administration in small to medium size companies in the UK have been lost, and very few of these companies are prepared to hire staff just to "look after computers" (many even baulk at getting maintenance contracts for the equipment they do have, just using commodity PCs and cloud services such as MS365 or similar)
@vfrmedia @mansr @bloor lost ? In many cases they never had them. Cobbled together IT held together by random people in the company who did computery stuff at home, and now and then bodged up by a consultant or two.
For a lot of businesses customer facing IT was always a non critical afterthought and always remotely hosted. It just moved from small co website/email hosting to big cloud aggregation.
And on a cost and returns basis totally rational not an underinvestment IMHO
For many its only recently become more critical due to the use of online banking and/or various types of cloud-based EPOS terminals as well as accounting software and staff scheduling services (which previously more often used to be run on local machines) - and its often not even that the business /couldn't/ go back to older methods, but they required more "slack" time away from the core business which smaller businesses no longer want to spare (especially as entry level jobs like office juniors have increasingly disappeared)
@penguin42 @beasts I would hazard a guess that the customers of @beasts are sufficiently clueful to not make them a single point of failure. The implicit rejection of big cloud and use of @beasts I think is a barometer of this.
@penguin42 @bloor most modern stuff seems to be engineered for MPOF (multiple points of failure). We once came across a static webpage that depended on five different CDNs for dynamically loaded resources.
@bloor I've never been in favor of having someone else manage my data. Call me possessive, but I put all of my important data on thumb drives.
@bloor I mean, solarwinds and crowdstrike are still commercially successful concerns, so clearly the threshold of business damage required to stop using a bad vendor is really, really, really high
okay i have to ask this. it has been driving me mad for so long now.
why is pizza in the UK so ridiculously expensive?
"pepperoni feast" at pizza hut UK (pickup): £20.99 (~A$42.50)
"pepperoni" at dominos australia (pickup): A$8 (~£4)
what could possibly cause this?? if you bought a pizza from dominos AU and shipped it internationally it'd probably cost about the same!
somebody will, of course, reply to tell me "ah, but there's coupons, you can get it for half price if you blah blah". sure. i accept that. however. that also exists in australia!! "you could get it for only £12" but you could get the aus one for only $4! (not factorial)
what is happening??? are the italians imposing a 300% pizza tariff on the UK as punishment for putting corn on pizza? is it thatcher's fault? is this why detective halligan has a pitzer tab? help
Alt...
pizza hut UK pricing. a medium pizza is £20.99; small is £17.99; large is £22.99. they do, however, dull the pain a little by putting an adorable little loop in the base of the pound character
Alt...
dominos australia pepperoni pizza for $8
Some time ago I posted my rules of DIY.
Rule 1 of DIY: if you're struggling, you're likely doing something wrong. Wrong tool, wrong method, wrong idea.
Not to be confused with the other Rule 1 of DIY: however long you think something will take you, bank on it taking at least twice as long.
And of course there's Rule 1 of DIY: some jobs are best left to professionals.
Now, I am not big on DIY, but I will engage in it when necessary. Also, my hobby has enough DIY aspects anyway, and I quite enjoy those. But I am fundamentally lazy, and that means when I do a job, I want to be able to do it with the least amount of friction and hassle.
So today, for no particular reason, I am going to add Rule 1 of DIY: look after your tools. Keep them clean and maintained. Otherwise they'll fail you when you most need them.
I've just received this text message. The names are correct and the sender's number matches the name, but the sender isn't somebody who would use this language/tone or even contact anybody this way at all - I don't believe this is anything other than a scam for a moment, but I'm struggling to see what the angle is here. What are the odds on me receiving another "please send money to..." or similar text shortly? Or does this scam work a different way?
![Hi all, my son and his Girlfriend had a really bad accident last week in their holiday in Portugal, [redacted] is bruised and battered and [redacted] is on life support after [redacted] pulled her from a burning car. Any help would be gratefully received as they have lost 90% of their belongings. I am with them in Portugal. If you could share this as wide as possible I would so appreciate it - thank you.](https://mastodon.me.uk/system/media_attachments/files/115/378/120/099/307/737/original/27b45f5351067c79.jpeg "Hi all, my son and his Girlfriend had a really bad accident last week in their holiday in Portugal, [redacted] is bruised and battered and [redacted] is on life support after [redacted] pulled her from a burning car. Any help would be gratefully received as they have lost 90% of their belongings. I am with them in Portugal. If you could share this as wide as possible I would so appreciate it - thank you.")
Alt...
Hi all, my son and his Girlfriend had a really bad accident last week in their holiday in Portugal, [redacted] is bruised and battered and [redacted] is on life support after [redacted] pulled her from a burning car. Any help would be gratefully received as they have lost 90% of their belongings. I am with them in Portugal. If you could share this as wide as possible I would so appreciate it - thank you.
@NAB I suspect they're waiting for you to respond in some way
@CatherineFlick @Floppy You're both probably right, but if that's the case, it does mean that the scammers must have stolen my contact's phone - which would also explain how they knew the name of their son and his girlfriend. I've sent a message to their partner to make them aware... we'll see what happens.
@NAB
My guess (no more) is that you're right - a request for cash will quickly follow. And they've got your number.
Curious about the details though.
@NAB I reckon you'll only get the link if you reply... otherwise they'd have sent it straight away. There's probably some people who would fall for this and reply that *would* be alerted by the inclusion of the link.
Confirmed as a scam **SEE BELOW** (as if we didn't already know). Interestingly, it seems that my contact still has their phone, so I wonder if it's a dodgy app or something else...
**EDIT** Not confirmed as a scam. Not confirmed as anything yet.
@NAB so weird. Sometimes they're those scams that are just fishing for any response, seeing if they can get anyone to even reply.
Oh great. Texted my contact's partner, letting them know what was happening. They told me they'd let them know. Just received another message from my contact saying "It was from me". Clearly their partner has just texted them with my message. 🙄
@NAB Maybe I’m being paranoid, but would their partner not know and be able to confirm what had happened?
Now this is just getting weirder and weirder.
Probably don't need to redact as much info as I did before, so...
Just had a reply from my contact's husband (who is also the father of the son in the original message):
"Hi Nicholas, have just spoken with XXXX and she did send this message to her contacts on her phone, she has also put a post on FB"
So, he's the father of the man in the car crash, his wife has flown out to Portugal and he never thought to mention any of this before.
Plus I sent him the text of the original message and he's not commented on that either.
I'm still thinking scam, but this has become very very weird. Maybe his phone was stolen/compromised too.
@nick None that couldn't be controlled by their phones (WhatsApp, FB etc.)
@NAB If this was me (in the position of the initial message sender) I don't know why I would be messaging my contact list asking them to spread this sort of news far and wide?
Is it actual SMS text messages, or from some kind of messaging app? I don't recognise the UI (but also I haven't had an Android phone in a long time).
@missiongiraffe I agree. Text message. I thought the same, but the number for both her and her husband match what I've contacted them on before.
@NAB I'm really invested in this now. I hope it is a scam because the real situation sounds awful, but I don't understand what they are gaining from this, unless it's going to lead to some kind of 'gofundme' type link at some point?
I wouldn't be worrying about anyone's lost belongings though when someone is on life support in a foreign country!?
@missiongiraffe And guess what.... There's a GoFundMe link.
Tried calling him. Slight pause before it rings and after a few rings, AI "Sorry can't take your call" message. Suspect calls diverted elsewhere.
@NAB It’s a lot of effort to go to this isn’t it, crikey.
A friend of mind had something similar, someone took over her phone number, they think by going into the relevant phone shop and passing off ID, who didn’t follow checks properly.
I wonder how many members of the house of commons would be able to pass an A level in English...
@quixoticgeek I’m pretty sure I couldn’t and I’ve been working in publishing for nearly 20 years. Good thing I have EUSS.
@Nicovel0 I grew up in the UK. I couldn't...
@quixoticgeek @Nicovel0 I grew up in the UK, with English as my first (and if I'm honest, only) language. I got a D at GCSE (E in English Literature). I'm pretty sure I couldn't get an English A-level. Not unless the standards have dropped a *lot* since I was at school (I'm not suggesting standards have dropped)
@quixoticgeek Disablist as heck too. We already had to backtrack on forcing everyone to pass English and Maths GCSE before they were permitted to stop trying age 18 and leave education cos it just didn't work.
This is another spiteful dogwhistle way of blocking people bringing their families to the UK with them. It's vile.
@nick @NatalyaD @quixoticgeek right?! Really not equivalent. Unless there's an English-as-a-foreign-language A Level.
And do they mean a full 2 year A level? Or A1/AS (I'm old)?
B2 is really only a small step up from B1.
@quixoticgeek just read the article and it says B2. Um.
Yeah no. It's not A Level. If B1 is considered GCSE... I mean yes B2 is a step up but not that big! And there's a chasm of a difference between English subject A level and English as a foreign language A Level (not even sure that exists?).
I mean, I don't think the Spanish language A level is equivalent to an English A Level.
I took a B1 German test to get citizenship. I could have done B2, but I really didn't fancy risking not passing.
@quixoticgeek they mean equivalent fluency to an A-level in Spanish or German, not the English A-levels that native speakers take. None of the news articles mention this, though.
You can't even get an A-level in English as a second language, it's just based on CEFR equivalence.
1-0. 💋
@Geri watching football?
@FediWouter yes
@Geri enjoy! xx
@FediWouter I am beginning to like it
But can think of better things to do
@Geri I am not so interested in football 😊. But it is nice to see people enjoying it!
@FediWouter Ray gets very excited
Some of that rubs off on me xx
@Geri Like yawning xx
@FediWouter yes he yawns I yawns
And the reverse is true
We are both Scorpios
We hate sprouts, Tories and Nutella
@neil tata consulting will no doubt claim that they already are, and that what they built for jlr coop m&s etc already was and yet…. ????????
@nick @neil it’s usually much more stupid than this
Put things on the internet
Sack (sorry… “rightsize”) any in house expertise
Award ongoing management to cheapest bidder, who either don’t get told to do a full audit or get told it’s too expensive and that the documentation is “definitely up to date”
Don’t update things facing the Internet fast enough (or at all)
Get pwned
Make shocked pikachu face and ask for government help, I guess
Heat Pump Saga update.
Octopus are coming to cap off the gas and remove the gas meter today. This will save us approx £9 per month in standing charges.
Meanwhile recognising that this isn’t a particularly scientific comparison it is nonetheless interesting to note that we have averaged about 12kWh electricity per day for heating and hot water combined this prior week. Meanwhile if we look at gas usage for the same week last year we averaged nearer 100kWh per day.
As rough gross error checks go, I think that’s enough for me to say it’s somewhat working. Now of course financially a kWh of gas is 6p whereas a kWh of electricity is 22p. But still
Alt...
Heat pump energy usage graph
Alt...
Octopus energy gas usage graph for week of 6th oct 2024
@bloor 637Kwh per week? We use less than 300Kwh combined per month for our 3 bed/3 storey semi occupied 24/7... What eats up so much power?
@rbairwell I suspect some of these factors probably play into it :
* big house
* detatched house
* windy/hilltop location
* boiler that was condemned and (water) leaking
* 15+ year old (guess) rather inefficient boiler
@bloor I'm quite impressed with the savings compared to the same month the previous year. (My heat pump was installed in June this year).
boostedThis property no longer burns stuff.
Alt...
Gas meter gone. Gas disconnected.
@bloor Awesome! I did that a few years back - bar a solid fuel stove. It's fantastic. One odd side-effect is that my house is quieter.
@bloor Oh the earth link between your un-gas internal piping and the external gas pipe is interesting.
I realised I didn't put actual numbers around the two weekly values.
So on the new system, heating and hot water are using 12.44kWh electricity per day average.
On the old system, this time last year, we were using 91.08kWh gas per day average.
Saying again weather could be different for sure, that is still 13.7% of the energy used.
Electricity 21.89p/kWh
Gas 5.97p/kWh
12.44*0.2189 = £2.72 a day elec
91.08*0.0579 = £5.44 a day gas
BUT on a cost basis... weirdly, exactly half.
It feels mad and slightly indefensible that you can use 13% of the energy, and still pay 50% of the money. But the good news is that at some date in the not too distant future, I'll be able to either buy overnight, charge batteries and use during the day. Or, ultimately, a good chunk of the year, generate my own entirely.
I see Octopus's electricity rates are going up in the unit cost, but down in the standing charge. I am currently on a "fixed anytime rate" because we have EVs but don't do heaps of miles, and we have a heatpump which potters along all the time. And a hot tub of sorts.
Current kWh = 21.98p
Current standing = 62.22p
New kWh = 25.27p
New standing = 43.66p
Insanely irritatingly, i looked at the new gas pricing... 5.71p/kWh so gas is actually falling whilst electricity is getting more expensive.
HAHA and - I now can't see via the Octopus app what my gas tariff WAS because at midnight Octopus now longer shows gas (because it got disconnected today!).
Ofc I could look at old invoices from them, so there is that.
@bloor this is why I need my own data system. We changed suppliers and lost ALL of our historical usage data. 😳🤬
@bloor same moving house - I lost access to my historical consumption data for the old house. Thankfully Home Assistant remembers.
@bloor Octopus's historical prices for all electricity tariffs are available via their API to anybody, whether a customer or not. Haven't checked but I think gas prices are, too.
@bloor If you can sort your LiFePO4 batteries out you could use their ultra cheap rate (not sure what it is, but a mate of mine has PV and storage plus a heat pump and EV and he's rather chuffed).
@davep Yep!
Planning to do just that with Fogstar 32kWh and a VIctron Quattro 15K.
@davep The current 4 hours nighttime rate is just 8.5p a kWh.
So I could fill up my battery for a maximum of £2.55 a day at that rate. And ofc charge cars at the same time etc.
@davep I think the Quattro 15K and pull 41A peak off the 240v mains and shove it into the 48v battery side, meaning around 200A on the DC side. So in theory it should easily be able to fully charge a 32kWh battery pack from dead in under 4 hours.
@bloor Should be pretty easy, but you'd only want that in winter really if you risk losing extra daytime insolation (unless you can sell it for more).
@davep My notional plan is to do without solar, initially. So just have a battery storage type system, with grid and V2L inputs, and an output into my house/garage.
Then once that is working nicely and reliably, then find a solar installer who'll just do the roofwork/rails/panels to isolators in my loft, and let me hook it all up myself.
My theory is, if they see a well installed and reliable setup already, they may be more willing to add solar to it, piecemeal.
@bloor @davep Octopus Go is 5 hours. Eon Next drive is I think 6 or 7. When I did our maths both cosy and next drive beat the standard tariff without batteries for our use. Back then Eon was 7 at 7p or so, which with heating tending to be night biased and water tank and some load shifting was an easy win, though we went battery anyway.
@nick need a compatible car or charger though, which I won’t have
@etchedpixels @davep my strategy is going to be to call roofers who can do solar rather than solar companies
@bloor I am on Tracker for gas, and the rates are usually pretty low - I think it would be hard to make a heat pump work for me financially. Which is a big problem - IMHO the charges that aren't directly related to electricity production/distribution should be moved onto just the gas supply to make electric heating a more clear cut choice. Some of this might improve as gas is removed from the generation mix I suppose, but that feels like an unnecessary delay.
@bloor I think it's fair; you're paying for a lot more infrastructure with the electric; for the gas it's just being pumped to you; for elec if it's gas burning it's being burnt somewhere and running through the turbine (loses a big chunk, so more gas goes in to get you that elec energy out), then also their is payment for the wind/solar infrastructure, and cablling and pylons and the supply of blue smoke.
@penguin42 hmmmmmm I guess so … intuitively though it just feels like gas infra is so much more of an ongoing infra liability. But you are right, it probably isn’t.
@bloor @penguin42 they’re putting new gas mains in on the street next to ours, right now.
Very defensible for people to stay on gas for the foreseeable future. More houses going onto gas than coming off, still in 2025 in the UK. Meanwhile, our gas heating was double the carbon of our beef consumption last year, which itself was 2.5x the carbon of our total ICEV miles. Our leccy was 1/10th the carbon of our gas heating, or zero with our Ripple.
@bloor Our only remaining gas usage is for cooking. At £9 a month it'll take 15 years or so for the cumulative standing charge to exceed the cost of a new cooker (it's an 1100mm double oven range).
Or is it the other way round, that it'll take 15 years or so for the cost of a new cooker to be amortised against the removed standing charge?
There's probably a sunk cost fallacy and some false economy in there somewhere.
Anyway, I'll probably be dead before then either way!
Ticket prices for next year’s #MikroTik Professionals Conference are reduced until October 31st. Early birds really do catch the worm as space is also limited for the VIP Dinner so please make sure you book your tickets early rather than at the last minute and find we’re sold out. See https://mtpc.world for more info. Also we’re now calling for presenters for MikroTik related tech talks. Got a great solution using their gear? Tell us about it and you could be presenting it in March! 😀 #MTPC
Was served a YouTube ad for some kind of small plug-in room heater – i.e. the heater is kind of built right onto the plug, like a wifi booster – that will apparently be in every home in the UK by next year and thought, hm, that seems unlikely, and a ten second research revealed that yes, they’re illegal and will probably burn your house down! If you don’t have a suspicious nature online, how the fuck do you survive?
This is a tad special of @jlcpcb - they altered placement for one of my parts.
What did they change? Well, they do not say, but there is a clue here - if you look closely this is not "straight", it is a few degrees off. I can assure you my placement was at 0 degrees rotation, dead straight. Looking the other side you can see the alignment pegs are not centre of holes - what I uploaded was, exactly.
So WTF did they change it to be "off" like this? I'm assuming it will end up straight.
Alt...
The image features a close-up of a circuit board with a prominent blue rectangular component overlay. Various labeled pins and markings, such as "DAT0," "DAT1," "CMD," and "CLK," are visible near the edges of the component.
Alt...
The image displays a circuit board labeled "Audio Recorder." It features various electronic components, including capacitors and connectors, with visible traces in blue and pink colors. The design emphasizes the layout for audio recording functionality.
@jlcpcb Ooh, and another clue.
They moved the component bottom right in first picture slightly (they don't say how much or why and it is tiny). They normally do that if too close to another component. Well it was not.
But having turned the SD card holder slightly, it may now be too close. That is mental!
@jlcpcb I am pretty much having to use crayons now to explain.
Alt...
Image of a circuit board labeled "Recorder" with annotations indicating specific components: "U1," "Exactly centre," and "As I ordered." Various connection points and design elements are visible in the background.
Alt...
An image of a printed circuit board (PCB) featuring a microcontroller. Annotations highlight that the component is rotated, no longer centered, and indicate reference points labeled as "on centre" and "image from DFM".
Alt...
A close-up of a circuit board showcasing labeled components. Key text highlights include "Recorder" and "No longer in centre," with additional annotations indicating "DFM image." The image includes various electronic elements and connections.
Alt...
An electronic circuit board featuring a white connector labeled "J4." There are pink dots indicating connection points, along with text stating, "The file I uploaded as shown on your order page."
I feel like randomly promoting a really amazing Free Software project:
Yes. Invidious is a self-hosted YouTube frontend. It provides a web interface for you to search/watch YouTube videos.
You can run it yourself, or use a public instance.
I installed *my own* private instance, on my network. It's working great.
Your browser won't run all the YouTube bloat. Invidious still has to run Youtube's challenge.
Really, really great project. Way better than YouTube Premium.
@libreleah YouTube Premium does have the benefit of giving each video's creator more money per view, if I remember correctly.
@krans donate $1 to the creator directly. that's more money than they'll ever get from youtube, if all their viewers did that. and not one cent goes to google.
@libreleah Serious question: have you tried doing that yourself?
@krans yes.
@krans when i like something enough, i pay.
@libreleah Okay, yes, I pay a number of people via Patreon, but I think that, “The recommender algorithm suggested this mini documentary about sinkholes in Dorset and I learned something,” level of videos also deserve financial support.
“Track down the PayPal of everyone whose video you watch,” is not a thing anyone can do in practice. Therefore I can only conclude that (1) using Invidious will deprive people of income for their work and that (2) you're fine with this.
@krans i think your central thesis, that invidious takes money away from creators, is ultimately wrong, and the same kind of moral panic argument used for decades against such technologies. you also miss the point entirely, that this is about taking control away from google. google gives creators peanuts. almost all the revenue goes to google. if we take your argument to its extreme, then you're basically saying we should watch youtube.com with ads turned on, or pay for premium. no thank you.
@libreleah @krans @neil for those starting out (who have reached the payout threshold) YT creation does rather depend on the model of “each viewer earns me something”. Youtube appears to have started not counting adblocked views as well, so now there is an algorithmic penalty for the creator. For all of G/YTs many faults, if “manually send money to things you like” worked at scale as a compensation model, more FOSS projects would be fully funded. 2c
@interpipes @krans @neil all of this is yet more evidence of why google is bad. they keep piling shit on users and expecting them to tolerate it. ads (read: spyware in your browser) should be removed from the internet, in their entirety. and should upload their videos to a peertube account, preferably self-hosted. the centralisation of the web is precisely what people like myself wish to avoid.
if a few people profit from google's abusive practises, the risk is theirs. nothing to do with me.
@libreleah @krans @neil sorry, but for me this comes out to “to take some $ out of Gs pocket, I’m ok with doing the same to a person who’s video I watched but didn’t like enough to pay for and it’s their fault because they didn’t have the resources/time/tech chops to pay up front for hosting their video even though it offers no compensation, dramatically hurts their discoverability & audience size, all while I admit YT has utility in using this shim”
@libreleah @krans @neil it’s fine to be idealistic about how you want the world to be but you cannot pretend this is consequence free in the world that exists for the people for whom youtube is probably the best way to get started because that is where the audience is
@nick aiui marketing like that is mostly about saturation / familiarity, so that when you see it later during an “opportunity to purchase”, it is not alien to you
![[?]](https://files.mastodon.social/accounts/avatars/000/284/255/original/180b5757a494b7a9.jpeg)
![[?]](https://social.v.st/system/accounts/avatars/109/360/452/395/342/558/original/ee08367ecd483af9.png)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/115/221/419/544/794/407/original/379eeece71440d86.jpg)
![[?]](https://files.mastodon.green/accounts/avatars/108/194/478/557/584/759/original/5f5b497e871aacfa.jpg)
![[?]](https://mastodon.nl/system/accounts/avatars/110/372/119/582/870/943/original/d9085a304286f8af.jpg)
![[?]](https://shore.me.uk/social/nick/s/cf6308305561d13738fb690e63e1c169.png)
![[?]](https://files.mastodon.social/accounts/avatars/109/199/285/835/114/515/original/e8300c6b54345d9e.png)
![[?]](https://social.v.st/system/accounts/avatars/110/663/121/091/154/845/original/a0014703fade6bd3.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/110/698/003/048/581/761/original/ab41d88abde947cf.png)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/109/293/499/907/445/807/original/b7cfc9a8b57a0087.jpeg)
![[?]](https://media.infosec.exchange/infosec.exchange/accounts/avatars/109/268/811/402/354/651/original/9aa5ccb755e0336b.png)
![[?]](https://files.mastodon.social/accounts/avatars/000/005/081/original/c43a9cab874507ac.gif)
![[?]](https://media.hachyderm.io/accounts/avatars/109/536/372/557/791/970/original/283ef1ac68fcb0ea.jpg)
![[?]](https://media.scicomm.xyz/accounts/avatars/108/193/011/536/398/715/original/dcb2083d500ee058.png)
![[?]](https://media.mastodon.scot/accounts/avatars/109/318/834/600/188/352/original/ac7b0c240f73ee70.jpg)
![[?]](https://media.hachyderm.io/accounts/avatars/109/285/155/241/586/065/original/261593f1871cd8cf.jpg)
![[?]](https://asset.circumstances.run/accounts/avatars/115/040/397/185/602/189/original/37ddb9aa92131b77.png)
![[?]](https://mangezdespomm.es/fileserver/01ZDA1VZ6EK7WZDECFCZAYZ74G/attachment/original/01K5HPE4GRFJFC12MBYZ32YZY3.png)
![[?]](https://assets.chaos.social/accounts/avatars/111/125/977/349/387/024/original/77158d1e334d6464.jpg)
![[?]](https://social-coop-media.ams3.cdn.digitaloceanspaces.com/accounts/avatars/109/943/451/572/743/095/original/f88785d8808b9d4d.jpg)
![[?]](https://beamship.mpaq.org/system/accounts/avatars/111/291/143/572/888/607/original/250162ac1088b753.jpg)
![[?]](https://s3.eu-central-2.wasabisys.com/mastodonworld/accounts/avatars/111/894/545/329/128/529/original/816e146816d93148.png)
![[?]](data:image/png;base64, iVBORw0KGgoAAAANSUhEUgAAAEAAAABAAQAAAACCEkxzAAAAUUlEQVQoz43R0QkAMQwCUDdw/y3dwEvsvzlL4X1IoQkAisKmwfAFT3RgJHbQezpSRoXEqeqCL9BJBf7h3QbOCCxV5EVWMEMwG7K1/WODtlvxAYTtEsDU9F34AAAAAElFTkSuQmCC)
![[?]](https://files.mastodon.social/accounts/avatars/109/246/966/188/953/497/original/aa70201aae0b0829.png)
![[?]](https://social-cdn.apcn.nz/accounts/avatars/108/202/950/871/447/752/original/9154cdbb47459605.jpg)
![[?]](https://furry.engineer/system/accounts/avatars/109/842/314/074/409/741/original/63d72129ce3c1ee6.png)
![[?]](https://media.mas.to/accounts/avatars/115/455/310/026/380/920/original/932a97c0b5951472.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/000/016/609/original/media.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/000/414/457/original/e6dae35163cb9f17.png)
![[?]](https://files.mastodon.social/accounts/avatars/114/696/834/910/684/493/original/b84cdaea43af7a1c.jpg)
![[?]](https://mastodon.samfira.com/system/accounts/avatars/109/300/076/986/606/097/original/33d76e553254296e.png)
![[?]](https://media.tech.lgbt/accounts/avatars/110/638/379/428/793/655/original/1c7529069b599af6.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/113/991/527/337/015/621/original/876bd7fb8c655dc4.jpg)
![[?]](https://media.exquisite.social/exquisite/accounts/avatars/109/491/199/486/260/631/original/fd8523256a635901.png)
![[?]](https://stockage.framapiaf.org/framapiaf/accounts/avatars/000/106/763/original/73a1613fdf975e57.png)
![[?]](https://files.functional.cafe/accounts/avatars/109/338/203/127/697/084/original/22c9de88631142b1.jpg)
![[?]](https://cdn.masto.host/rubysocial/accounts/avatars/000/093/360/original/920f5a8782200f9b.jpg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/109/286/430/443/380/988/original/35a36e437a1b36e4.png)
![[?]](https://mastodon.neilzone.co.uk/system/accounts/avatars/000/000/003/original/1686a2608fa9766e.jpg)
![[?]](https://media.bsd.cafe/bsdmmedia01/accounts/avatars/110/740/169/477/916/693/original/dfa8f1b9beefa405.png)
![[?]](https://assets.chaos.social/accounts/avatars/111/009/267/204/974/574/original/1ff55e560f75efcb.png)
![[?]](https://cdn.fosstodon.org/accounts/avatars/000/297/041/original/16e82706ce6700c9.png)
![[?]](https://media.bsd.cafe/bsdmmedia01/accounts/avatars/111/697/277/820/456/122/original/886dd177470837e8.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/109/768/883/930/305/553/original/887f2caba9eb0333.jpeg)
![[?]](https://runbsd.duckdns.org/system/accounts/avatars/115/044/211/226/145/314/original/997fedc68eea9550.jpeg)
sysop ![[?]](https://assets.growers.social/accounts/avatars/111/020/532/060/811/975/original/e282b2b6ba9f6a6c.jpeg)
![[?]](https://media.mstdn.social/accounts/avatars/110/423/678/736/047/689/original/53d7ab22ea23351d.png)
![[?]](https://files.mastodon.social/accounts/avatars/114/588/214/663/585/912/original/6239b72426979054.png)
![[?]](https://mastodon.opencloud.lu/system/accounts/avatars/000/002/330/original/438b72f2c764dcfe.jpg)
![[?]](https://media.bsd.cafe/bsdmmedia01/accounts/avatars/112/333/060/238/284/470/original/1c072481f6f94ff9.png)
![[?]](https://files.mastodon.social/accounts/avatars/107/408/823/620/736/413/original/b570c4e978e49a10.png)
![[?]](https://files.mastodon.social/accounts/avatars/112/431/607/410/187/418/original/01dc616584c8e09f.jpg)
![[?]](https://mikrotik.social/system/accounts/avatars/110/383/270/751/993/877/original/64e5ca41b8c3a427.jpg)
![[?]](https://duckpon.de/fileserver/01KCA9MAPTK4NJ7TA8B3QEFQRF/attachment/original/01HRX03CDCFVR43XE0GFX1X9W8.png)
![[?]](https://social.huginn.uk/system/accounts/avatars/110/595/369/111/728/209/original/43f9563f372046f4.jpeg)
![[?]](https://assets.chaos.social/accounts/avatars/112/582/837/077/852/364/original/a95c87cdc45171ed.png)
![[?]](https://cdn.masto.host/mastogaloophcom/accounts/avatars/000/000/001/original/18a9f4eed57df90d.jpg)
![[?]](https://cdn.masto.host/twitsocial/accounts/avatars/109/295/725/014/989/802/original/e2202c05edbf0ef5.jpg)
![[?]](https://know.me.uk/system/accounts/avatars/114/638/593/337/995/648/original/1d6b185daabf53f4.jpg)
![[?]](https://media.dragonscave.space/accounts/avatars/109/326/405/195/348/480/original/a2b09fceaebaa89d.jpg)
![[?]](https://media.swecyb.com/accounts/avatars/114/114/408/086/738/353/original/d60eca397a1e3c1b.png)
![[?]](https://media.mas.to/accounts/avatars/109/253/089/914/451/486/original/a8de8587cb88328b.png)
![[?]](https://fedi.duckduckpigeon.co.uk/fileserver/01H7PZ308AG4XW3YAR9PNCZ6VD/attachment/original/01JPW84WG5S1A6FKPRB66XKGTY.png)
![[?]](https://comam.es/snac/susie.png)
![[?]](https://snac.bsd.cafe/teajaygrey/s/2d49dbf4b3c8576152b35aef1cffac46.gif)
![[?]](https://gone.lema.org/fileserver/01AGG3YT2F4M3S9YGAKRQAP768/attachment/original/01K2QJ616KA8K531FV56FRH7A3.png)
👾![[?]](https://media.darkdork.dev/ce/b9/2e/ceb92e36d41ee669de1191d2d22c6e1f615aa15c3775a4e68fc9c34aeb51a3a8.jpg)
![[?]](https://assets.chaos.social/accounts/avatars/000/273/333/original/55e33c6abd01fec0.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/000/301/292/original/5e9b33fbc7c81cc4.png)
![[?]](https://social.crablab.uk/system/accounts/avatars/111/675/941/783/517/422/original/ab99458adba57a6e.png)
![[?]](https://media.beige.party/accounts/avatars/113/367/594/452/099/022/original/94861ff7a73fe253.jpg)
![[?]](https://social-coop-media.ams3.cdn.digitaloceanspaces.com/accounts/avatars/109/298/017/786/958/083/original/de2a7aa347726f69.jpg)
![[?]](https://media.mastodon.scot/accounts/avatars/109/932/887/982/577/246/original/a99c1b1e4d88c0cd.jpeg)
![[?]](https://files.functional.cafe/accounts/avatars/108/197/412/836/833/295/original/36fd30acd388fccb.gif)
![[?]](https://cache.treehouse.systems/accounts/avatars/112/099/706/214/118/070/original/fbb9e788e2b9dfa5.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/114/612/855/535/960/576/original/916a30a3fcf36df6.png)
![[?]](https://media.mathstodon.xyz/accounts/avatars/109/293/519/649/435/368/original/01885f8a02807a4f.jpg)
![[?]](https://cdn.fosstodon.org/accounts/avatars/109/325/399/667/863/822/original/53afdc234f26de81.png)
![[?]](https://media.mstdn.social/accounts/avatars/109/349/941/912/741/020/original/8dc30e25f99a6669.jpg)
![[?]](https://mastodon.radio/system/accounts/avatars/109/366/667/985/541/784/original/18b54cf63883c0c2.jpeg)
![[?]](https://cyberplace.social/system/accounts/avatars/113/431/345/677/238/539/original/23b4e257d54a32a2.png)
![[?]](https://media.mas.to/accounts/avatars/115/012/022/939/128/676/original/95ccd06b722719d5.jpeg)
![[?]](https://cdn.mastodon.org.uk/accounts/avatars/000/000/458/original/11730f2dd8d95b54.jpg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/110/119/576/402/736/812/original/db37cb6d93e5a242.jpg)
![[?]](https://static.mstdn.fr/static/accounts/avatars/000/017/118/original/1c72f0dd033841cb.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/000/040/107/original/753d82117a1d8149.jpg)
![[?]](https://mastodon.durrans.com/system/accounts/avatars/113/341/704/915/711/585/original/145b095a311c5f73.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/113/180/523/570/931/154/original/944a80db8d6e8ec2.jpg)
![[?]](https://stockroom.wandering.shop/accounts/avatars/109/245/282/271/507/015/original/fcd4dae226f5f8bd.png)
![[?]](https://know.me.uk/system/accounts/avatars/112/135/188/733/562/337/original/90471e203f13080d.jpeg)
![[?]](https://martinh.net/system/accounts/avatars/109/400/880/109/060/310/original/5172e3328444ce3c.gif)
![[?]](https://social.mitexleo.one/system/accounts/avatars/115/271/953/765/819/112/original/19a815118e11e122.jpg)
![[?]](https://assets.chaos.social/accounts/avatars/000/244/507/original/c2951c2a27d722fa.png)
![[?]](https://cdn.masto.host/bloortw/accounts/avatars/109/270/268/145/356/367/original/819ce26120a92c4f.jpg)
![[?]](https://morehammer.uk/system/accounts/avatars/109/350/000/993/890/285/original/eb5504c0ccaf5faf.jpeg)
![[?]](https://toot.hoyle.me.uk/system/accounts/avatars/109/314/228/859/843/431/original/f3b424c884fb0f30.jpg)
![[?]](https://mastodon.garyhawkins.uk/system/accounts/avatars/109/264/901/804/483/365/original/b7f44f3d4eb659c4.png)
![[?]](https://files.mastodon.social/accounts/avatars/000/027/421/original/7b875d3654b7d343.png)
![[?]](https://files.mastodon.online/accounts/avatars/110/203/472/458/921/791/original/ebea493dcc3a16bb.jpg)
![[?]](https://media.mstdn.social/accounts/avatars/108/194/335/102/996/817/original/bb94cd67d14981fd.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/110/549/421/416/780/169/original/fcc5a44057c64a33.jpg)
![[?]](https://media.hachyderm.io/accounts/avatars/109/359/148/639/098/596/original/77a87a844cc7c97e.png)
![[?]](https://cyberplace.social/system/accounts/avatars/109/557/327/504/609/477/original/ee7f1278616c8aa9.png)
![[?]](https://files.mastodon.social/accounts/avatars/000/260/057/original/433fb295062a91d2.jpg)
![[?]](https://go.mxtthxw.art/fileserver/01VJDFF4AJ69WHRRCQFGT4XH2Z/attachment/original/01K3DSZ0BS4K2F39NMV4RBMNR3.jpeg)
![[?]](https://media.beige.party/accounts/avatars/113/189/588/106/176/292/original/0e38607fb2ec1609.png)
![[?]](https://mastodon.acc.sunet.se/system/accounts/avatars/109/296/533/593/186/082/original/90ce886e138d6388.jpg)
![[?]](https://anarres.family/system/accounts/avatars/115/378/737/612/000/119/original/b6c11aae3c57dc3f.jpg)
![[?]](https://fedi.slightly.tech/fileserver/01AW7Q631K31VGK5SXTZM50SFD/attachment/original/01K1K1RBNFPZNEMDJ93JYAK3HC.png)
![[?]](https://media.infosec.exchange/infosec.exchange/accounts/avatars/109/290/344/195/278/226/original/9f6e794cb5787d02.jpeg)
![[?]](https://cdn.periwinkle.nl/Mastodon-System/accounts/avatars/113/112/834/201/175/939/original/4791966176e9a480.png)
![[?]](https://media.infosec.exchange/infosec.exchange/accounts/avatars/109/290/296/993/925/843/original/5be47c8c77400509.jpeg)
![[?]](https://cdn.masto.host/sauropodswin/accounts/avatars/109/300/578/355/719/354/original/9db26c12c87e5a62.png)
![[?]](https://gts.sadauskas.id.au/fileserver/01F7Z9N32BK120W7P9PERCP98E/attachment/original/01JS6QN8YVQZP33VKFRTV7V2DD.jpeg)
![[?]](https://files.mastodon.green/accounts/avatars/109/285/226/650/634/456/original/bff14963d3189323.png)
![[?]](https://media.beige.party/accounts/avatars/111/170/451/699/072/582/original/bab2b237388663c2.png)
![[?]](https://media.mas.to/accounts/avatars/108/226/963/747/243/077/original/57e217f7bcc86a2a.jpg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/108/205/804/885/085/404/original/7f50c3fa5744002b.png)
![[?]](https://media.infosec.exchange/infosec.exchange/accounts/avatars/113/849/437/066/281/241/original/fb2cf6a8a9de3012.jpg)
![[?]](https://media.mastodon.scot/accounts/avatars/113/487/421/532/296/259/original/561b4064f5fb3326.jpg)
![[?]](https://cdn.mastodon.org.uk/accounts/avatars/000/079/094/original/e36b82ae7b4c7f96.jpg)
![[?]](https://s3.masto.ai/accounts/avatars/109/309/646/333/981/670/original/25d2f2e6f27ce761.jpg)
![[?]](https://media.mas.to/accounts/avatars/109/279/230/934/902/973/original/bf24c5f969e97a3d.jpg)
![[?]](https://media.mathstodon.xyz/accounts/avatars/000/000/013/original/d3bb832f7c670bec.png)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/109/285/502/679/731/083/original/f62a52cd45faa914.jpg)
![[?]](https://stockroom.wandering.shop/accounts/avatars/110/346/882/732/049/066/original/cd2e0fcaf0bdaea2.jpg)
![[?]](https://cdn.masto.host/appdotnet/accounts/avatars/108/200/003/332/597/568/original/ddc404d7c40a4810.png)
![[?]](https://media.hachyderm.io/accounts/avatars/109/534/322/998/122/893/original/240034a5301a9617.jpeg)
![[?]](https://a.mindlycdn.com/accounts/avatars/109/388/287/375/823/721/original/57d7637698176cd4.png)
![[?]](https://cdn.masto.host/poorlyrenderedcom/accounts/avatars/109/854/939/558/130/302/original/c1505fddfda6ba64.jpeg)
![[?]](https://files.mastodon.social/accounts/avatars/109/240/161/492/241/732/original/c99a4f8947d2b8ab.jpeg)
![[?]](https://files.mastodon.social/accounts/avatars/109/251/322/665/847/225/original/425b5ddd7c092039.png)
![[?]](https://s3.masto.ai/accounts/avatars/109/340/908/805/254/753/original/ca11583f8b6287ce.png)
![[?]](https://cdn.social.linux.pizza/system/accounts/avatars/109/288/463/804/828/991/original/e570523759a732c6.gif)
![[?]](https://files.mastodon.social/accounts/avatars/000/223/608/original/03b5f1a25b0a5830.jpeg)
![[?]](https://s3.c.im/accounts/avatars/109/309/536/306/748/095/original/5b9868de41c560c5.jpg)
![[?]](https://cdn.masto.host/toastdragon2611net/accounts/avatars/110/582/794/509/985/886/original/4651fd28c050d7c3.jpg)
![[?]](https://cdn.mastodon.org.uk/accounts/avatars/109/276/772/405/158/471/original/c4fe528a4c77fe90.png)
![[?]](https://cdn.cupoftea.social/accounts/avatars/110/752/150/558/293/084/original/8c4a05f6f77c7704.png)
![[?]](https://media.hachyderm.io/accounts/avatars/109/303/344/607/564/678/original/e7f382748e1a42f0.png)
![[?]](https://thx.gg/system/accounts/avatars/109/310/653/474/331/897/original/615152febe3488e9.jpg)
![[?]](https://social.hodgson.io/system/accounts/avatars/109/350/106/916/313/114/original/8c7136e7cd35d4b8.jpg)
![[?]](https://toot.ale.gd/system/accounts/avatars/109/796/852/286/747/908/original/c9a0d62a6c2296f6.jpg)
![[?]](https://perfect.moe/system/accounts/avatars/109/344/742/183/127/883/original/c1c96826996fa09f.png)
![[?]](https://files.mastodon.social/accounts/avatars/109/246/078/948/794/953/original/13e2ecd309465b55.gif)
![[?]](https://society.oftrolls.com/system/accounts/avatars/000/017/356/original/a333546b7198014f.jpg)
![[?]](https://f2.tchncs.de/accounts/avatars/000/005/899/original/e25b9214df0b7bc5.jpg)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/109/287/626/612/367/826/original/afaec12ebb38ef71.jpeg)
![[?]](https://social.mythic-beasts.com/system/accounts/avatars/109/303/541/678/580/907/original/aca29febd934fece.png)
![[?]](https://files.mastodon.social/accounts/avatars/113/759/515/866/908/040/original/31668ffaceb627a0.jpg)
![[?]](https://files.mastodon.social/accounts/avatars/113/109/435/248/988/723/original/1893a349d2a56858.webp)
![[?]](https://wasabi.lynnesbian.space/the-slum-bucket/accounts/avatars/000/000/002/original/d1636e816ab84c6d.png)
![[?]](https://mediacdn.aus.social/accounts/avatars/115/307/910/016/340/275/original/49c103a3826bb5b8.png)
![[?]](https://mastodon.me.uk/system/accounts/avatars/109/264/670/017/475/803/original/7eb37a8aaeb05e67.png)
![[?]](https://mastodon.me.uk/system/accounts/avatars/109/273/831/845/008/925/original/a448d8b8fe181a37.jpeg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/000/000/001/original/12d6a44966a2920c.jpeg)
![[?]](https://s3.c.im/accounts/avatars/109/308/711/020/000/246/original/71b7781c839b3b1f.png)
![[?]](https://mastodon.me.uk/system/accounts/avatars/108/582/663/341/900/948/original/d6ca3941781b05ea.jpeg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/109/399/185/014/264/863/original/8a43eee85d23e01f.jpeg)
![[?]](https://files.mastodon.social/accounts/avatars/000/913/783/original/57afcddbe121a0b3.png)
![[?]](https://files.mastodon.online/accounts/avatars/108/316/080/764/868/660/original/7998e8a468ca85d1.png)
![[?]](https://stockroom.wandering.shop/accounts/avatars/109/740/670/253/239/218/original/3bb20e4d40f6a318.jpeg)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/109/285/593/194/158/221/original/186f500195a41d27.jpg)
![[?]](https://assets.disabled.social/accounts/avatars/109/327/284/425/145/650/original/6035c9243111557a.png)
![[?]](https://cdn.masto.host/frontendsocial/accounts/avatars/109/295/906/487/756/020/original/4d848189db38cb45.jpg)
![[?]](https://assets.chaos.social/accounts/avatars/000/285/148/original/dc079baa2f06fae3.jpg)
![[?]](https://s3.eu-central-2.wasabisys.com/mastodonworld/accounts/avatars/113/500/230/725/361/926/original/401da01dfee9507c.jpeg)
![[?]](https://files.mastodon.social/accounts/avatars/109/248/313/552/922/878/original/3c70ee29903b0413.png)
![[?]](https://files.mastodon.green/accounts/avatars/109/291/493/730/800/274/original/b2a6293ae30bfe42.jpg)
![[?]](https://cdn.masto.host/mastodonie/accounts/avatars/109/295/677/666/503/653/original/51a636f3f2b71390.jpeg)
![[?]](https://mastodon.nexusuk.org/system/accounts/avatars/109/326/346/926/693/431/original/41c477b79d528beb.jpg)
![[?]](https://media.infosec.exchange/infosec.exchange/accounts/avatars/000/001/831/original/d69e265d80b0946c.jpg)
![[?]](https://cdn.fosstodon.org/accounts/avatars/109/273/073/780/702/685/original/060af47e709ebec9.jpg)
![[?]](https://media.hachyderm.io/accounts/avatars/109/837/295/295/432/118/original/85a426f9975220b0.png)
![[?]](https://cdn.masto.host/mastodonart/accounts/avatars/109/837/099/322/388/566/original/04dfc0e2c250ae24.jpg)
![[?]](https://toot.me.uk/system/accounts/avatars/108/197/040/384/377/848/original/2069780af8f27343.png)
![[?]](https://media.mas.to/accounts/avatars/000/259/583/original/7470955940c84a59.jpg)
![[?]](https://mastodon.me.uk/system/accounts/avatars/110/491/786/726/464/127/original/be433b57dffe6484.jpg)