492 karma · joined November 5, 2015
https://www.generativestorytelling.ai/
@thedevlinb
www.github.com/devlinb
meet.hn/city/us-Seattle
Socials:
- github.com/devlinb
- x.com/thedevlinb
---
If the issue impacts tens of millions of customers, then yes, get it fixed right now. Extended outages can be front page news. Too many in a row and people leave the service.
Ideally monitoring catches outages when they first get started and run books have steps to quickly restore service even if a full fix cannot be put into place immediately.
I commented that my hospital network has the ability for me to see test results, communicate with medical providers asynchronously, and lets healthcare providers communicate what I said to them with each other.
Your counter proposal was a notepad and a pen.
Those two things have very different feature sets.
A site not directly hooked up to the main hospital system where I can add notes of things I want to talk to during my next visit, sure, maybe that is a decent proposal to trade off security vs convenience.
> We should reduce the security of the entire healthcare system because you can't adult?
Do you expect patients dealing with depression or anxiety to be able to keep track of that notepad? How about patients who don't have secure housing, or who live with an abusive partner?
Or anyone who is elderly and just forgets things now and then.
Or anyone who is not neurotypical and has issues with memory?
At my child's pediatrician I can upload images through the web portal and have a nurse call me back anytime of the day or night. If there's any follow-up questions at my child's next appointment, his doctor has full access to all communications that happened through both text message and the web portal.
This kind of 24-hour digitally connected healthcare access with a huge boon as a first-time parent and made life a lot easier, especially when incidents like when my son woke up at 3:00 in the morning screaming and then projectile vomited 6 ft across his room (which by the way, was nothing to worry about and apparently completely normal... So long as it just happened once.)
List of medication side effects, dosing guidelines and so forth have been common throughout the industry almost since it's very inception. Indeed, there are books going back thousands of years across multiple cultures around the world that are just referenced guides for medical practitioners.
Scheduling an urgent care appointment is connected to my account at the hospital network. When I step into the hospital and get tests done, everything is automatically uploaded to a web portal where I can view it and doctors can easily forward my test results to other facilities. Lot of the imaging work is actually done at third party facilities but they still show up in my medical records , presumably having been forwarded.
When my appointment begins my doctor can look up any comments I left when scheduling the appointment so she knows why I'm there.
Is it possible all the different buildings and facilities that are part of the hospital Network I belong to which extends across multiple counties across my state could all be running on their own private isolated network that is air gapped with medical records manually transferred over to the web portal via sneaker net?
Sure, but no one is going to do that.
> The sad reality is that companies don't want easy and efficient customer service... As that would allow customers to cancel their continued payments...
This is not true at all. The true reason is that a well trained call center employee can easily cost a company $20 per customer issue resolved (total cost inclusive of training, office space, equipment, etc).
For any low margin business (e.g. hardware under $500 USD!) that basically destroys the entire profit from that customer.
Customer service is expensive.
Think Microsoft in the late 80s through the 90s.
If you are charging for the platform, you can let thousands of other businesses try the risky hard to scale stuff and you'll get a cut of everything no matter who wins.
That typically didn't happen because engineering reviews had to occur first.
A single command created a new repo, setup ingress/egress configs in AWS, and setup all the boilerplate to handle secrets management, environment configs, and the like.
All that was left to do was business logic.
Outbound sales is being automated.
Lots of very hum-drum stuff. Ordering room service at a hotel for example.
Tons of data entry jobs are now gone.
LLMs are already better at humans from a cost perspective for many tasks.
Which means calculating how long free going to take for a given object is incredibly difficult. Which is my entire point. Of course it's not non-deterministic unless someone 's destructor call s some random code but it is not easy to calculate and in the real world is basically unpredictable.
Spitting up a separate thread and doing an unknown amount of work is no more predictable than doing on the main thread.
Fundamentally destructors can do whatever the hell they want which means you never know how long destruction is going to take.
And of course if you're freeing up an object that has a linked list inside of it and you need to clear out the link list, now you're jumping all around memory and that's never fun, and the run time sure as heck is not constant based on what needs to be paged in and out and what exists in cache lines where.
I'm not saying these problems are unsolvable for a given use case and there are good reasons why custom allocators abound throughout the industry, My main point is that manual memory management does not necessarily lead easy to predict run times for freeing up memory.
This especially true because a lot of developers think that malloc and free are just magic somehow.
In reality, garbage collectors often aren't that complicated, and when it comes to large complicated user applications, you're going to be spending a lot of time in the allocator and deallocator no matter what memory management schema you choose to use.
(Edit: and then there's fragmentation which is the death of many manually managed memory systems! A naively Java or c-sharp application can run for far longer than a naively written C++ application!)
Micro services solve a problem for companies that have already reached a certain scale were cross team communications have become unfeasible and now communicating by well defined API contract is a better choice.
Also ideally micro services should reduce the blast radius of outages. If an outage is not easily traced to what team is root causing it, then proper monitoring is not in place.
Sure I've had times where on call went off and it wasn't my team's fault but it was no more than 10 minutes to determine that and reroute the call and then to back to sleep.
The other fact is that when designing a new microservice it should be done in conjunction with the primary consumers of that service just like designing any other part of software. Stakeholders need to be brought in and consulted.
The advantage of microservices is that new code is only going to impact direct consumer s and downstream services.
It also allows you to upgrade a service in place or completely rewrite it so long as you adhere to the original contract.
I've seen impressive rollouts of security updates across a huge code base that was only possible because of a microservice-based design.
I've seen giant monoliths fall apart as multi-year long efforts are undertaken just to update the build system.
The hard part of a microservices is they require discipline in an organization and basically assigning an engineer for 1 to 2 weeks to write run books and add monitoring.
Of course, those runbooks should be written no matter what paradigm someone uses!
At HBO Max every incident had a full writeup and then real solutions were put in place to make the service more stable.
My team had around 3 incidents in 2 years.
If the cultural expectation is that the on call buzzer will never go off, and that it going off is a Bad Thing, then on call itself isn't a problem.
Or as I was fond of saying "my number one design criteria (for software) is that everyone gets to sleep through the night."
The customers win (stable service) and the engineers win (sleep).
I've worked on tiny embedded systems (.net micro framework) where for a given usage pattern the GC was perfectly predictable, as it should be.
When NTFS came out it was way better than anything on Linux. Heck even in 2006 NTFS was better.
But Linux keeps getting new file systems while Windows keeps NTFS.
Nvidia is making boatloads of money right now selling GPUs to companies that think they will be making boatloads of money in the future.
Nvidia has the better end of things at this very moment in time.
They basically showed it was possible to get a machine at that lower price point and then capitalism and mass market manufacturing did the rest.
Is this so often the case innovation requires someone to prove what's possible and then going down the same path is much easier for those who come later.
I remember when I first tried to SketchUp I was horrified at how atrocious the UI is compared to rhino 3D.
Keeps you alive at 30, kills you at 70. You got to reproduce so it counts as a win for evolution!
CL had a miserable mobile experience up until a few years ago, no good mobile app, and spam has always been a problem on the platform.
Localization is hard, even for companies that spend a lot of time and effort on it.
It isn't just string replacements!
In America, before the Internet took off, every year everyone would get a book called the "white pages" that had the name, address, and phone number, of everyone who lived in their city.
The American view of privacy is that "openness makes for a civil society".
Although one can argue that hasn't been working out well for us lately ..
Likewise, marriages are publicly recorded and accessible online, as are all property purchases, births, deaths, and even property tax payments.
Though for some reason we consider income taxes to be super secret. Everything else is public, but not those! (How much cash someone put down to buy a house? Public. How much money that person makes? Not public. How much money everyone donates to politicians? Public.)
Apple owns the stack and iterates slowly.
Most laptops change to some of their chips (wifi, graphics, etc) every year. But it can easily take a couple of years of experience working with a shop and writing drivers for it to fix every power usage bug.
For many years (!!) Microsoft actually kept the same wireless chipset in their Surface machines. Was it the latest and greatest? No, but it worked!
Then or course you have software. For years Firefox had a bug where if any open tab has GPU accelerated content, it would prevent the entire machine from going to sleep.
If Microsoft clamps down on that sort of behavior it'll break someone's software (complaints about laptops going to sleep fall into 1 of 2 camps, it happens too often or it doesn't happen enough !)
Until there is no one left to buy the actual product, at which point those on top suffer as well.
If I have a multi leg journey I'll make sure every other stop is at an Airbnb with a washing machine.