Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[nenjin]

Welcome to Software As A Service babyyyyyyyyyyyyyyyyy. My company has shifted almost entirely in the last year from "we prefer if you host your own server for our services" to "we prefer if we host you."

Although this seems better suited to the WTF, Mildly Upset threads no?

The NRA has been denied its claim of bankruptcy, because... they are not bankrupt.

Note that their motion was dismissed without prejudice. Which means they can refile the same motion a different way with additional materials, most likely to be considered by a different judge, and still get the move approved. A dismissal with prejudice would have been an absolute shutdown of the attempt.

[McTraveller]

Biden (or someone) still needs to fix health care for all its other reasons.  My family's premiums are > $10,000 a year (my part + my employer's part).  That's what we pay even if we have zero health care expenditures.  I could have picked a plan that had lower premiums but a higher deductible, but that ends up actually costing more if you need care!  ($4800 in out-of-pocket premiums; with $2000 deductible, versus ~$5500 in premiums with $500 deductible.)

Read that again:  we pay more than $10k a year just so we can spend more money if we actually have a health care need.  Ok technically we have zero out of pocket for preventive annual physicals, but even those end up not being $0 because often they run labs (cholesterol, metabolic panels, etc.), and those labs are not free.

Salesforce, a major service provider for businesses, has been completely out for more than an hour now. Looks like a national issue.

Our services in many areas have ground at an absolute halt, we deliver everything from food to toys to medicine. A lot of businesses are going to be impacted today.

These computer-related outages are getting scary.  Even more scary than COVID, if you ask me.

[Telgin]

Much like the AWS outage that meant people's doorbells and vacuum cleaners didn't work anymore.  We're all putting more eggs in a single basket than we even realize.

[nenjin]

Which is why smart devices and smart homes have always seemed like a bad idea to me.

[Dunamisdeos]

Companies like Amazon use salesforce. I'm super surprised this isn't on the news.

We're gonna see some stock changes, probably.

[anewaname]

Liz Cheney spoke in the House session today. Here is part of it...

Spoiler (click to show/hide)

THE TRUMP DEPARTMENT OF JUSTICE INVESTIGATED THE FORMER PRESIDENT'S CLAIMS OF WIDESPREAD FRAUD AND FOUND NO EVIDENCE TO SUPPORT THEM. THE ELECTION IS OVER. THAT IS THE RULE OF LAW. THAT IS OUR CONSTITUTIONAL PROCESS. THOSE WHO REFUSE TO ACCEPT THE RULINGS OF OUR COURTS ARE AT WAR WITH THE CONSTITUTION. OUR DUTY IS CLEAR. EVERY ONE OF US WHO HAS SWORN THE OATH MUST ACT TO PREVENT THE UNRAVELING OF OUR DEMOCRACY. THIS IS NOT ABOUT POLICY. THIS IS NOT ABOUT PARTISANSHIP. THIS IS ABOUT OUR DUTY AS AMERICANS. REMAINING SILENT AND IGNORING THE LIE EMBOLDENS THE LIAR. I WILL NOT PARTICIPATE IN THAT. I WILL NOT SIT BACK AND WATCH IN SILENCE WHILE OTHERS LEAD OUR PARTY DOWN A PATH THAT ABANDONS THE RULE OF LAW AND JOINS THE FORMER PRESIDENT'S CRUSADE TO UNDERMINE OUR DEMOCRACY.

The part in bold reminded me of this James S.A. Corey quote...

Spoiler (click to show/hide)

“Desperate psychotic people do desperate psychotic things when they’re exposed. I refuse to grant them immunity from exposure out of fear of their reaction. When you do, the desperate psychos wind up in charge.” - from "The Expanse" by James S.A. Corey

[Frumple]

Companies like Amazon use salesforce. I'm super surprised this isn't on the news.

We're gonna see some stock changes, probably.

What was on the news is whatever that is that's going on with that pipeline. Something about a cyber attack/ransomware attempt shutting down a major fuel line? Any idea if it's related?

[Vector]

Companies like Amazon use salesforce. I'm super surprised this isn't on the news.

We're gonna see some stock changes, probably.

There's a whole frigging tower locally (in SF) that you couldn't ignore if you tried.

It feels in some way like the era of magical US supply chains might be over, at least for now (see: toilet paper and yeast shortages). This might be for the best though, I guess.

[MorleyDev]

Thing with the more centralized infrastructures is the overall productivity/capabilities/efficiency is flat-out better when it's working than if everybody rolled/hosted their own, but more things are impacted at once when it goes wrong.

I'd actually bet that if you measured the overall economic cost/growth of the centralized SaaS (including the rate of outages) to the rate of localized outages when everyone is rolling/hosting their own the localized ones would be more expensive to businesses/economic growth overall, but since it's always localized to the one business nobody else notices so it doesn't feel as significant.

[McTraveller]

If only there were some other metric aside from pure efficiency...

I mean in many systems we sacrifice efficiency for reliability and robustness.  In fact, we even require it for things like aviation, some medicine, civil engineering, etc.  You'd think we'd require it for infrastructure (gas pipelines, water, power), including de facto infrastructure (major financial services).

[MorleyDev]

What I mean is, I suspect that the most optimistic scenario for the infrastructure being re-decentralized is the only difference is say, to simplify the numbers for explanation, rather than 3 banks going down for a day every 3 years, you'd have 1 bank going down a day every year.

So it's not really a loss, it's just that everyone notices when the 3 banks go down whilst only a 1/3 will notice if one of them goes down.

More likely (sticking with the 3 imaginary banks), those 3 banks probably can offer more services using the third party and you'd probably have 1 bank down for at least two days a year if it was in-house. So you actually gain uptime overall.

The centralized infrastructure is usually more stable than the homegrown stuff, since the businesses can consistently invest time and resources into it and they typically do have SLAs* with their clients wherein they pay fees if they don't meet a certain percentage of uptime. It's also the only way smaller businesses can have access to some things they couldn't reasonably develop/host themselves (e.g SMS messaging services).

* A 99.9% uptime does still mean they can be down for 8h 45m 56s a year before crossing that threshold.

[Dunamisdeos]

Companies like Amazon use salesforce. I'm super surprised this isn't on the news.

We're gonna see some stock changes, probably.

What was on the news is whatever that is that's going on with that pipeline. Something about a cyber attack/ransomware attempt shutting down a major fuel line? Any idea if it's related?

We discussed it as a possibility but the actual cause has been super hush hush. I wouldn't be surprised.

[Iduno]

https://twitter.com/chriswhoisgreat/status/1391867835123748865

More comments of "unskilled" labor, which they can't find enough people both skilled enough and willing to risk themselves for the pay.

If only there were some other metric aside from pure efficiency...

I mean in many systems we sacrifice efficiency for reliability and robustness.  In fact, we even require it for things like aviation, some medicine, civil engineering, etc.  You'd think we'd require it for infrastructure (gas pipelines, water, power), including de facto infrastructure (major financial services).

Civils put in the infrastructure, which is sometimes the problem. I know around here, instead of taking into account the heavy rains we frequently get, they use the average annual rainfall, leading to flooding issues a few times per year. Because that's the standard.

As long as you don't consider maintenance costs (capital costs come from a different budget), it technically costs less in exchange for roads/water/power/etc. not working for several days/year.

[McTraveller]

Right but real robustness is if you have those 3 players, and any customer of one can instantaneously switch to another one if one goes down. So you don't have 3 independent suppliers, you have 3 redundant suppliers.

But redundancy costs a lot more than just having enough for average.

@Iduno yes, that is the issue - do you build for "common" or do you build for "exceptional."  Exceptional costs a lot more, but it's like you're paying for insurance up front all at once.  It's a difficult problem to solve; I'd rather have AI trying to figure out a better cost model for infrastructure or other systematic disruptions, rather than just looking at dollar cost / discount models which are based on guesses about interest rates and rates of return anyway.

[dragdeler]

-