A couple of weeks ago, YouTube went down for a few hours. Over 240,000 users in the US hit Downdetector to confirm they were not alone in their suffering. For a brief, beautiful moment, the internet collectively remembered that it has no idea how to do anything without a tutorial video.
It came back up. Everyone went back to what they were doing. Lesson not learned.
I get it. YouTube being down is a minor inconvenience for most people, and I am not here to catastrophize a few hours of buffering frustration. However, there is a useful conversation buried inside that outage, and since we are all apparently still here and operational, now seems like a good time to have it.
When “Down” Becomes an Actual Problem
Here is the question I ask clients before something goes wrong, which is the only useful time to ask it: What would actually stop your business if it went offline today?
Maybe not YouTube. But let’s work through the real list.
Your email. If Microsoft 365 or Google Workspace has an outage, and both have had them, what happens to your team for the next four hours? Do you have a backup communication method, or does everyone sit and wait?
Your practice management or billing software. For accounting and law firms, especially, these are the operational spine of the business. If your cloud-based platform goes down on a filing deadline day, what is your plan? If the answer is “we would figure it out,” that is not a plan.
Your VoIP phone system. Most modern phone setups run over the internet. No internet, no phones. If a client is trying to reach you during a crisis and your phones are tied to the same connection that is down, you have compounded the problem.
Your client portal. If your clients cannot access shared documents or submit information through your portal, and that is how you have set up your workflow, a platform outage becomes your problem even when the fault is entirely upstream.
The Thing Nobody Talks About Until It Is Too Late
There is a concept in IT called a single point of failure. It means any component in a system that, if it fails, takes everything else with it. Good business continuity planning means identifying chokepoints and either building redundancy around them or making a deliberate, eyes-open decision to accept the risk.
The problem is that most businesses have never done that exercise. They built workflows around tools that worked, and never asked what would happen when those tools stopped working. Which they will. If there is anything technology is reliably good at, it is failing at the absolute worst possible time.
I am not saying you need a redundant system for everything. That would be expensive and largely unnecessary. I am saying you should be able to answer the question: if this goes down for four hours on a Tuesday, what do we do? For some tools, the answer is “nothing critical, we wait.” For others, the answer should never be “we have not thought about it.”
A Starting Point That Does Not Require a Consultant
You can do this yourself in an afternoon.
List the five tools your business genuinely cannot operate without. Not the ones that would be annoying to lose. The ones that would actually stop work.
For each one, look up the vendor’s status page. Every major platform has one. Bookmark them. This is where you check during an outage instead of refreshing your own screen for twenty minutes.
Then ask, for each tool: if this were unavailable for four hours on a deadline day, what would we do? If you do not have an answer, that is the gap. It does not have to be a complex solution. Sometimes the answer is a shared cell phone number and clear instructions for staff on how to communicate.
Test your backup communication method before you need it. If your email goes down, does your team know how to reach each other? Do they have each other’s phone numbers? Does anyone know where the physical contact list is, if one exists?
The Real Lesson from YouTube Being Down
The reason the YouTube outage got attention is that it hit something people use constantly and rely on emotionally, not operationally. The outages that actually cost businesses money are quieter, less covered, and affect platforms most people outside your industry have never heard of.
Microsoft had a significant outage in January 2025 that affected Teams and Outlook. The coverage was minimal compared to the YouTube story. The operational impact for businesses that rely on those tools daily was considerably worse.
The platforms you depend on to run your business will go down at some point. Planning for that is not paranoia. It is the same logic as keeping a spare tire in your trunk. You are not expecting to use it. You do not want to find out you need one when you are already on the side of the road.
If you would like help mapping out your firm’s actual operational dependencies, that is a conversation worth having before the next outage, not after.
Quick and Easy: YouTube’s February outage was mostly harmless, but it is a useful reminder that every platform your business depends on will go down eventually. The consequential ones are not the entertainment platforms. Taking an afternoon to map your five most critical tools and ask “what do we do if this is unavailable?” is more valuable than any uptime guarantee a vendor has ever offered you.
You probably already knew this: YouTube is the second most visited website on the internet. In obvious first place is Google.com which also happens to be the parent/sister company of the world’s biggest video streaming site. YouTube has over 800 million videos (and growing) and gets over 17 billion visits per month (source), so saying they make a lot of money on that website off your eyeballs is putting it very mildly. The secret sauce, of course, is the algorithm that keeping must-see videos constantly into your viewing experience, and because it’s Google-powered, you can bet those engineers know exactly how build a data-driven, personalized algorithm that knows exactly what you want to see. Or does it?
One Algorithm to rule them all?
Based on the platform’s success and profitability it’s pretty clear that this algorithm is doing something right, but there is still plenty of criticism and scrutiny on YouTube’s content selection, especially in light of continuing misinformation problems plaguing all social media platforms. If you are a user of YouTube (statistically likely!) you are probably already familiar with the various tools you can use to supposedly tailor YouTube’s algorithm to only provide content aligned with your interests. There are even buttons to dislike, remove from recommendations, or report as misinformation, but according to research done by Mozilla Foundation (full disclosure: a non-profit research and advocacy organization that is funded by Firefox money and search engine royalties from Google, etc.), these buttons are essentially ineffective. My takeaway? YouTube is using the age-old marketing trick in offering the illusion of control, while still driving traffic to the videos and trends that make them the most money. The article is lengthy, but Mozilla helpfully provides an infographic summary that is a bit easier to digest and leads to the true reason they published these findings. In the end, Mozilla is an activist organization attempting to drill some transparency into the biggest content platforms. The only way this is going to happen is if enough people step up and ask for change. You don’t have to stop using YouTube, but recognizing their placebo controls might give you better insight into why true control over your feed feels elusive.
Image by Pablo Jimeno from Pixabay
Previously I wrote about the Elephant on the Internet, and lately it seems like we can’t stop blundering into the pachyderm that shall not be mentioned. Last week, Medium published a controversial article about a strangely mutated (but inexplicably popular) genre of kids videos on YouTube. For those of us hardened by years of work (and play) in the darkest and weirdest corners of the internet, the article wasn’t surprising, but it was definitely disturbing how bad things had become in this area. If you don’t mind wearing the mental equivalent of hip-waders, James Bridle’s article plays Rod Serling to this Twilight Zone-esque subgenre that evolved to exploit YouTube’s keyword and “Suggested Videos” algorithm. One of my “favorite” videos from this story is entitled, “BURIED ALIVE Outdoor Playground Finger Family Song Nursery Rhymes Animation Education Learning Video”. Rolls right off the tongue, eh?
What this means for you
A few years back, my wife and I made the sad (but not surprising) discovery that YouTube was not something that could be left in a child’s hands unsupervised. At the time, it had yet to grow the strange and mutated mushrooms that crowd the darker corners as described in Bridle’s article, but we encountered too many inappropriate “suggestions” from YouTube’s algorithms and came to the conclusion that (a) nobody was driving this particular bus, and (b) some people would do anything to make a buck, especially if they could do it by exploiting technology. In other words – not family friendly, and definitely not kid safe. A few years after that, Google announced YouTube Kids – a walled-garden subset of age-appropriate content that parents could trust to entertain their progeny, and we had a brief glimmer of hope that someone at Google noticed their space needed some adult supervision.
It’s no secret that children’s content is an evergreen but highly competitive industry. Prior to the internet, media companies would spend millions chasing short attention spans in the hopes of cashing in on an ephemeral merchandising craze, eg. Cabbage Patch Kids, Tickle-me Elmo and Baby Einstein videos. Now, thanks to the popularity of crowd-generated content, YouTube is a top destination for Internet “Gold Rushers” with children’s videos a particularly profitable and exploitable “vein”. The problem is not with the creators of these freaky videos – capitalism and Internet make for some strange, but predictable bedfellows. It’s that YouTube is yet another example of a system that has gotten away from its creators, and despite their attempts and promises to close yet another Pandora’s box, the sheer size and scale of the Internet continues to overwhelm and surprise the companies that laid the groundwork for its current dominance.
To sum up: it should come as no surprise that when the Internet gets ahold of something and everyone’s too busy watching the scenery to drive the bus, we can end up on the wrong side of town with no idea how to get back. Add YouTube to the crowd of monsters (Twitter, Facebook, Equifax, Wikileaks etc.) that have gotten away from their masters in service of agendas outside of their control.
Image courtesy of TAW4 at FreeDigitalPhotos.net
On October 26 of last week, a number of popular, “cloud-based” services suffered multi-hour interruptions. Among the outages was Google’s App Engine, a platform that is used by thousands of other websites and internet platforms including one of my favorites, Passpack.com. Some of your favorites may have been impacted as well: Dropbox, Tumblr and even YouTube were affected. For many, this was a non-event, particularly those who operate and compute within enterprise-based platforms, or rely solely on the desktop and storage of their own computers. C2 Technology relies heavily on cloud-based services, primarily Google products, for our core information systems, and I use Passpack to track the multitude of passwords I need to do my work. So when those outages hit on the 26th, I found myself unable to access the keys to my various digital kingdoms, and felt very much like someone who finds themselves locked out of their car, and at the mercy of another person’s timetable. In this particular case, Passpack.com wasn’t even to blame, as their own reliance on Google’s App Engine service hamstrung their ability to deliver service to their customers, and the fine engineers at Google themselves were struggling with the outage. Everyone’s brand took a hit, and yet there was no one any one of us could blame for the outage – not even a radical hackivist group looking to ruin someone’s day for political currency.
What this means for you:
Very simply, “Never put all your eggs into one basket.” This homily, however pastoral-seeming, still very much applies to how you should use technology, especially when it comes to your core business processes. As an illustration of how this can be bad: I was using Passpack to store my Gmail password, which was complicated and impossible to remember, and instead relying on a complicated, but easier-to-remember passphrase to access Passpack to retrieve that password whenever I needed it. When Passpack went down, so did my ability to access Gmail and all of my client contact information. The lesson to take away from this: if you are going to store critical information online, have a back-up plan for continuing to operate without access to that information. Either back-it up locally (fraught with its own set of risks), or compartmentalize parts of your operations so that they aren’t heavily reliant on a single service provider, or the presence of the internet.
Image courtesy of “vichie81” / FreeDigitalPhotos.net
Google’s New Advertising Shill: You!
MetaFilter user Andrew Lewis coined a phrase that has become the rallying cry for internet privacy watchdogs over the past 3 years, “If you are not paying for it, you’re not the customer; you’re the product being sold.” He was speaking of Digg’s redesign in 2010 in which the emphasis of the site shifted away from user-centric content curation and towards a model that was clearly intended to monetize Digg’s large userbase. Since then, the phrase has been applied to many services, including the 800-lb gorilla of free internet services, Facebook, and dozens of other social media sites that use advertising money to fund their “free” services. Savvy users will note that Google has been leveraging this model on a less obvious (but no less profitable) basis ever since Google search arrived and Gmail extended its tendrils into millions of users’ daily online existence.
The subtlety was cast aside boldly last week when Google announced a change to its privacy policy that granted itself the right to utilize its users’ likeness and content authored on any one of its many properties to advertise to other users. This includes content and reviews written by users on G+, YouTube, Zagat, and the Google Play store. The new policy is the default, and users must opt out if they prefer to not participate in this endorsement model. Clearly, Google is hoping to entice advertisers with the very real impact of recommendations made to users by people they know. But many are angered by this change, and the internet outrage is spreading.
What this means for you:
If you have a Google account, then you are automatically opted in to this advertising model. To opt out, you must go to your Account settings under the Google+ section, and look for the “Shared Endorsements” link to disable your participation in the program. If you actually go do this, you’ll note that Google has written quite the argument as to why you might want to stay opted-in: “Your friends might not be able to benefit from your wisdom.” Depending on your level of participation in online reviewing/commenting/rating, participating in this program may be no big deal, or a very big deal. Either way, you should consider the implications for your online brand, whether current or planned, and the impact on your privacy, especially if your face and words could start appearing on thousands of monitors around the world.