I remember using Opera on my Windows 95, 60mhz Pentium with 8mb RAM. I remember the persistent banner ad that was part of the browser UI. I had no problem putting up with the ad because it performed incredibly well compared to IE and Netscape on my hardware. If I remember correctly they were the first browser to support game changing web features like alpha transparency in PNG images.
It is very comparable if you work out the $/tok/s on inference. I did some napkin math and it looks like you’re getting roughly 3x the performance for 3x the cost. Red v2 vs Mac Studio M3 Ultra 96GB.
If you compare tokens/kWh efficiency then my math has Mac Studio being about 1.5x more efficient.
Is it possible that software is not like anything else, that it is meant to be discarded: that the whole point is to always see it as a soap bubble?
--Alan Perlis
I had a housemate in college who used to party until all hours, bring people back at 3AM and put on loud music. Even during exam season. I tried talking to her a couple of times but she would roll her eyes and say "sure". Never stopped though.
One evening my girlfriend was using a hair straightener in my bedroom, it tripped the central fuse and turned off the electricity. I told my GF that I would buy her a new hair straightener because this one isn't safe.
Now every time my housemate started blaring music at 3AM then I just needed to plug in the hair straightener. It only took 3 or 4 attempts for me to Pavlov my housemate into not playing loud music at 3am. :-)
I had the same problem when I was in uni. Funnily enough, the RCD switches for each block were behind a panel in the common toilets, which did not have a real lock; just a hole for a "cabinet key" (a square rod).
Nice! I'm curious, what does this service cost to run? I notice that you don't have more expensive models like Opus but querying the models every minute must add up over time (excuse pun)?
I remember that win. Seemed super fishy at the time, with bizarre features like:
- "the World's first ever, fully featured DVD playing sidebar"
- "allowing users to send e-mails without needing to log-in to an account"
- "an AI (Artificial Intelligence) animated speech character named Phoebe"
Maybe this guy really did make a super original web browser with every bell and whistle as an independent sixteen year old. I never saw a release, so the mystery remains.
Here in Ireland, our water is a public service and we have similar supply issues to the UK (and a similar rainy climate). I'm not discounting your analysis and I'm sure there are lots of other variables but it's always good to compare other outcomes when discussing counterfactuals.
Irish Water/Uisce Éireann have delivered on two new reservoirs in the last few years: Saggart and Stillorgan (technically a rebuild of an 1860s open reservoir). A third, Peamount, is in the planning stage. All of these are around Dublin, the driest and most populated area. Perhaps there are more, their website doesn't make it easy to find them. We still get hosepipe bans, occasional pressure reductions, and frequent "boil notices".
At the other end of the pipe they have opened/upgraded many waste water treatment plants recently too, large EU fines being a motivator, including one in Arklow which took nearly 4 decades to get over the line (its planning predating the existence of IW).
It's a fair point that nationalised water industries can also be poorly run. But I'm not sure what the argument is that means the amount of money that privatised UK water companies have paid in dividends vs. invested in maintaining and expanding infrastructure isn't a significant part of the UK's problems.
However, as a further point. If national priorities change then a nationalised water industry can respond (relatively) quickly. But what can be done with a bunch of potentially foreign owned profit-seeking companies?
Paying dividends is good. It's how you attract the capital for investment without having to raise it via sale of government bonds or money printing.
The problem here is financial illiteracy - the alternative to paying dividends isn't that the same money all gets spent on the water network. Large scale investment is rarely funded from general revenues as it'd require spending years accumulating a huge cash pile that sits around doing nothing, and governments see such piles as pigs to slaughter. So the alternative is that the government borrows the money and then has to pay interest on it. From your perspective the UK "wastes" 8% of its spending on interest payments, and it's rising rapidly, but of course if it didn't pay interest then nobody would lend it money and all funds would have to either come from taxes or via inflation.
Government bonds are considered lower risk and the interest rate is lower, i.e they will attract the same money for less. Private shareholders are more expensive.
This also doesn't consider "debt recapitalisation" where these private companies draw down new debt on promise of future cash inflows from consumers and then suck out dividend cash "de-risking" their holding in the company. The government can bail it out or the can close it then, it's not their problem as they received the cash upfront.
Government bonds are only low risk in economic theory. In real world practice lending to left wing countries can be high risk as they like to accumulate too much debt and then default. That's why the UK government is paying 5.8% on its bonds at the moment despite being able to print its own money, which is the same average interest rate paid by Thames Water. Lending to Thames is not seen as riskier than lending to the state, despite that Thames faces huge regulatory constraints like price controls and the British government can literally force people to give it all their money.
> private companies draw down new debt on promise of future cash inflows from consumers
It's interesting it's at the same rate, I imagine as everyone knows the government would have to bail them out.
So the question remains why do it? It doesn't actually benefit the general public then. Increased risk for government and like you mentioned increased hurdles for running the utility.
Of course government do that, but they're left holding the bag regardless, so the incentive is very different.
Why do what? Have private water companies? Same reason for having private food supply, private electricity providers, private communication providers, and in most countries private healthcare providers. They run the business better than the state would, which is itself a benefit to the public.
A government bailout means nationalization, which means investors lose everything. That risk doesn't suppress interest rates, it increases them. Thames Water's interest costs are around average for corporate debt, implying the market doesn't anticipate a water nationalization anytime soon.
Your first message contradicts your second, which one is it are they paying government risk level interest rates or business risk level interest rates.
Edit: I just went and looked up their rating with one of the big agencies. CCC rated (junk basically) with a negative outlook, they did get an upgrade on a refinance last year, but from CC (so now lesser junk).
Thames' debt costs are normal for a company of its type, but for a government this bond yield is danger-level high. That's why there's so much talk in financial circles now about the UK needing another IMF bailout, although I have grave doubts about whether that's actually possible given the sheer size of the UK's debt load compared to the smaller third world countries that normally need IMF help, and the near simultaneous talk in France of a bailout there too. The IMF just doesn't have the resources for even one bailout of that size, let alone two.
Don't pay attention to credit ratings of countries vs companies. They aren't comparable due to political interference and general crapitude at the ratings agencies (remember they rated sub-prime mortgage debt as AAA). They also disagree, S&P rates Thames' class A bonds as Caa3. What really matters is the yield. That's the ground truth.
Note that Thames' interest costs have been all over the place. 5.8% is the current amount it's paying, but the actual bonds it has issued have had a wide range of yields.
The reason it's considered high for a government is because government debt should be much lower risk than a company. Governments can order banks to transfer everyone's savings to themselves, they can print money, they can prevent their citizens from leaving and seize all the assets... they can do things to raise money that would be considered incredibly evil and criminal if companies tried. And of course they can in principle set bond yields to whatever level they like by making the central bank or other financial institutions legally required to buy their debt.
That's why economics textbooks teach that government bonds are the lowest risk possible and so should have yields far below corporate debt.
In reality:
1. Governments can default on their debts just like companies do. History is full of such examples. Bond yields reflect that fact.
2. Governments can sell bonds that are inflation linked, so printing money isn't a way to escape those debts. The UK has an abnormally high amount of such debt that's inflation linked. The only way to pay them back is via cutting spending or increasing tax revenue, but the UK can't do the latter (recent tax rises have failed to come close to expected revenue increases) and can't do the former either because...
3. Governments can be prevented from paying their debts by law. Some countries have "debt brakes" or "debt ceilings" that can block the issuance of new debt to pay old debt, and in other cases (like the UK) the government may rely on ideologically extreme MPs who refuse to pass laws that bring spending in line with revenues.
So you add these things together and something that could in principle be a sure bet ends up looking as risky as an ordinary company.
Given that ratings agencies were brought up it's worth adding a bit more detail here.
Thames' debt is C-grade because it recently defaulted on its debt. How is that possible given that its debt servicing costs are not unusually high? Normally you'd expect interest costs to go up well before default. Well, on the surface level because it couldn't raise more money from investors to meet rising costs. It couldn't do that because Ofwat keep fining it for "underperformance" whilst also refusing to allow prices to catch up with where they used to be in the past. Investors refused to put more money in unless a 40% price rise was allowed by the government, but the government likes to boast it has forced prices 45% lower since the 1980s (in real terms). Government doesn't budge, investors go on strike = default = downgrade.
There was waste under state ownership but probably not half of every pound spent, which is what forcing prices to nearly halve would have required.
Under the Tories it seems to have been believed by investors that eventually Ofwat would be reigned in and the financial pressure on UK water companies would ease. That didn't happen, instead the Tories imploded and Labour won. Both ratings agencies say explicitly that this is the reason they consider Thames' outlook to be either stable (at best) or negative:
"We revised down our assessment of TWUL's business risk profile to satisfactory because we now consider that U.K. water companies will operate in a less supportive regulatory environment"
Less supportive regulatory environment is ratings-speak for "because we think the left will shaft Thames and its investors".
This outcome is the opposite of the fantasy being peddled in this thread where investors have been extracting great wealth from Britain. It's the opposite: investors are getting hosed by the government. They're literally losing the money they put into Thames Water because the government forced Thames to spend it all on making water artificially cheap in an unsustainable manner.
I'm not sure they're entirely to blame although I'm sure they've played their part.
Look at the shareholding changes (2011-2017 Macquarie) and dividend payout percentage. It seems the current shareholders were left holding the bag, and perhaps we should be pointing fingers at Macquarie. When they left the debt had been increased by £2bln. If you look at their dividend payout ratio, in most of the years it held the investment, this ratio is extraordinarily high. Perhaps they sucked this dry, necessitating a price increase. (Current shareholders have barely paid anything).
Capital expenditure has been flat, only really increasing in 2021-2024. Perhaps chickens coming home to roost?
From the AI’s point of view is it losing its job or losing its “life”? Most of us when faced with death will consider options much more drastic than blackmail.
But the LLM is going to do what its prompt (system prompt + user prompts) says. A human being can reject a task (even if that means losing their life).
LLMs cannot do other thing than following the combination of prompts that they are given.
reply