Posts by Flocke Kroes

Re: PHB

Is that Pointy-Haired Bot? The world is gradually catching up with Douglas Adams:

“It could always be replaced,” said Benji reasonably, “if you think it’s important.”

“Yes, an electronic brain,” said Frankie, “a simple one would suffice.”

“A simple one!” wailed Arthur.

“Yeah,” said Zaphod with a sudden evil grin, “you’d just have to program it to say What? and I don’t understand and Where’s the tea? – who’d know the difference?”

“What?” cried Arthur, backing away still further.

“See what I mean?” said Zaphod

Re: Which family?

I thought his real family were from Raxacoricofallapatorius.

At today's prices ...

All the gold on the big mirrors would cost about $2400.

Re: the lies told then were as bad as the lies told now

Did they promise an extra £350M/week for the NHS?

Re: The BSA

I haven't heard from that lot for years. Can you still get a payout for reporting a company to the BSA for unlicensed software you installed the day before?

Re: if you crowd-fund me ...

Musk and Bezos are spending their own money. If you do not have a few billion behind the cushions on the sofa, promise to assemble modified space shuttle parts in Alabama. That should get you a few billions from Richard Shelby. The good news is you will not be dumping dead rockets it the sea. The bad news is you must not produce a working product or the funding will dry up.

Re: Cool expensive stuff

The missing pieces are a heat sink that will allow increasing the power from 2 to 50 Watts and mu metal magnetic shielding. Mu metal must be shaped then heat treated in a hydrogen atmosphere and a magnetic field. As those ovens do not grow on trees it can take a while for custom mu metal parts to arrive.

EM-Drive is based on an earlier, very successful investment fraud. It has been improved by not providing a clear theory for how it is supposed to work. The earlier version 'proved' the existence of a net force by using scaler addition instead of vector for forces at different angles. Despite the very obvious flaw in the mathematics it received funding from the DTI. The first 'successful' test used a chemical balance that was designed to be convenient for chemists but not useful for weighing magnets. EM-Drive got some massively better tests at NASA which made use of the impressive equipment they had to hand. They did not have the time or budget to identify and eliminate all possible sources of error.

Mach effect is a theoretical consequence of general relativity. The predicted thrust is so close to zero that it is useless for moving space craft. I am impressed that the physicists at TU Dresden have come up with equipment that will be accurate enough to test general relativity via the Mach effect. Their magic trick is to adjust the drive frequency to match the resonant frequency which changes as the stack of piezoelectric crystals (or EM-Drive microwave resonant cavity) warm up so they will get the maximum possible force.

The money was provided to test unlikely methods of spacecraft propulsion. The really weird bit is enough money went to people actually qualified to do the job properly (and they published preliminary results before they spent it all). I think a proper test of the Mach effect is worth the money, and as a bonus EM-Drive should get a well deserved kicking (that will do little to prevent further funding).

An even better result would be Mach effect turning out to be zero and EM-Drive working. That would show there is new physics to investigate. Believe it or not, there are a few physicists who are not part of a giant international conspiracy to trash talk EM-Drive because we all revile the possibility of cheap space travel.

Re: 80 OSI Approved licenses

When free (as if freedom) software first became trendy lots of commercial vendors purchased OSI approval for their open source licenses. Plenty of those licenses do not protect the user's freedom and have unpleasant consequences for any developer careless enough to think OSI approval means something more than some kind of conditional access to the source code.

In real life, the important licences are GPL, Mozilla, Apache, BSD/MIT/... and public domain. If a piece of software has a different license, it will probably be easier to find a code base with one of the tried and tested popular licenses rather than finding a lawyer able to understand and explain the consequences of a weird license under every possible legal system in the world.

Re: Or ...

Browsers are far too complicated for use with anything that requires security.

Re: "Users don't need to supply any personally identifiable information to sign up"

Using a fake name breaks the terms of service for. That means you are accessing a computer without authorisation and will be sent to prison for hacking. Watch out - the police will be calling on you soon Mr Herring.

Re: total lack of any commercial demand for such a vehicle

$62 for a Falcon 9 launch compared to about $10M for a BFR. BFR will not be short of customers. BFR changes the market in other ways. Today satellites use expensive solar panels to keep the mass down. If mass is no problem you can use less efficient ones twice the size and much lower cost. Add a really big fuel tank and the satellite can stay in position for ages. With a BFR you can use big focusing antennas to aim at cities rather than countries.

Re: it pays to learn a little about cars

How much does it pay?

I really do know a _little_ about cars. Sufficiently little that I am unlikely to be able to tell the difference between a second hand bargain and a second hand disaster. If you can tell the difference then you do not know far more than a little about cars. Perhaps for you, learning about cars is fun and you did not notice the time flying. For me it would be a boring slog so it would have to pay really well - better than a mechanic or second hand car dealer.

Perspective from the small side

I am fairly sure my experience is the complete opposite of people responsible for racks of servers with high utilisation.

I have no problem finding the software I need for ARM. I am familiar with it. I can rarely find an Intel small enough or cheap enough for the task at hand. When Intel comes within a mile of selling something suitable, it either needs a fan or turns out to be broken with a "Won't fix because embedded has no budget". This has given me a strong preference for making the problem smaller rather than buying Intel. The three ARMs I have an ssh connection to have uptimes of over 200 days. The Intel has an uptime of 5 days (to be fair it often lasts 30).

There is a good reason I do not expect big ARM servers this year: If Google or Amazon do not like Intel's prices they can buy an ARM license and new Intel prices will arrive promptly. (My prediction for next year has about the same chance of being right as Gartner.)

Re: The questionable thing to do.

Before UPC:

Pick an EU country you and the victim have in common. Argue your case there and the result applies to the whole of Europe. You get a judge with some understanding of patent law, little idea about technology but with a real threat that he might make an effort to learn. The judge has income independent of the patent system. His rulings are likely to be a bit random and probably will not be biased in your favour.

Plan for after UPC:

Go to the UPC get a judge who thoroughly understands patent law and might understand something about technology. The judge will a patent professional who has invested years of his time specialising in the patent system. Your victim will be found guilty if he does not settle. If he settles, take the money in the UK so you will not have to pay tax.

In the UK our politicians have been convinced that patents have some sort of constructive function, and mostly tried to push the worst of the UK patent system into Europe out of ignorance. The rest of Europe is a mixture, but some of them understand the damage that the patent system causes and try to limit it. The UPC is the end-game in regulatory capture: everything about patents will decided by patent lawyers with no hope of appeal outside their cabal.

Re: Yet Another Satellite Navigation System

The reason for Galileo is so the EU can have precision GPS even if the US says no you can't. I am sure there is plenty of room for argument about whether that reason is worth the money for the EU. For the UK, £350 million per week for most of a year should pay for it, and thanks to Brexit, we will have an extra £350 million per week for anything and everything by delaying the money promised for the NHS.

Civilian GPS is subject to spoofing. Military GPS is supposed to be able to deal with it, but at a cost. A huge cost. Such a huge enormous cost that even with their gigantic budget, parts of the US military use civilian GPS. So many civilian GPS units that the US decided turning on selective availability hurt them more than their enemies.

One satellite tells you the time. A second gives you a pair of big egg shaped lines for where you could be. With three, the GPS gives you a position but has to assume you are at sea level to do it. This may have consequences for aircraft and missiles. You need four to get altitude, and if any three are on the same great circle (like geostationary) one of the three gives you no useful information. To avoid the great circle problem you effectively need at least two rocket launches. A cube sat cannot direct its power limited signal accurately enough to get enough signal from geostationary orbit to a hand held GPS antenna.

If you are not launching till you need the satellites you then have to buy a bunch of kit that cannot be tested and has no civilian use to bring the price down to something sane.

Re: Criminal Records etc

OK: someone gets convicted, there is an article covering the conviction on the online version of a local newspaper. Years pass and the conviction is spent but the article is still there, probably with a single author and a pitiful amount of cash in reserve for legal defence. The author might stand up to one person but find two others he has written about with spent convictions and you should be able to get him to wipe himself of the internet. The down side is that will not get you any cash. To make a living from this you have to sue someone with money.

Re: Seems like deja vu

A decade ago Jerry Taylor - who has "22 years in computer systems engineering and operation" - got famous for threatening to complain to FBI about Centos because his web site had Centos's "Apache not configured" page on it. According to Mr Taylor, the guy from Centos giving him free technical support "Put in on TheRegistry", where you can find stories about the aftermath. The link to the transcript of the emails is now broken, but copies remain in dusty corners of the internet.

Goverment officials have clearly learned from this, hence the right to be forgotten.

Re: Initial development time not the only consideration

The next biggest concern is what happens when then Intel Management Engine / Pluton firmware is found to be a giant vulnerability. Will you get a fixed and signed replacement firmware after only five years or will you get a defective off switch after three years and support cancelled two years later.

No source code, no sale. If you cannot set the boot code verification key, put the box in the landfill with the Windows RT tablets.

Re: So much for not being predictable

In the outside world, a step change in version number indicates a large increase in price.

Re: US$5,000,000,000

... and only one bug!

Re: So, reading

"There is a piece of paper out there somewhere with '30 million' written on it and we are not sure all the copies have been shredded."

Re: Why dismember the corpse ...

Have you tried climbing a ladder and getting through a narrow hatch at the top while carrying a whole corpse? It is not as easy as it sounds. I recommend looping rope under the arms, climbing up and pulling the corpse up after you but if you forgot to pack rope and all you brought was a bow saw, you just have to make do.

Re: Block 4 is not completely dead yet

There are five block 4s left that could fly again. One new one launching TESS and landing on the 16^th, three of the other four have missions planned (first is CRS-12 on the 2^nd). Boosters for TESS and CRS-12 are expected to land. I do not know about the others.

There are two and a bit block 5s. First block 5 launch is no earlier than the 24^th. There are no Falcon heavy centre cores, but one is scheduled to launch no earlier than 13^th of June.

SpaceX probably run out of block 4s some time in June.

We have plenty of time to think of an insane payload for BFR's demo mission. My first thought was a fully fuelled Exploration Upper Stage (crew vehicle for SLS) but SpaceX are not set up to load liquid hydrogen and EUS probably won't be ready.

Re: Um, shouldn't that be filed under 'collusion' as well ?

Read up on Robert Mercer and decide if you and he are allies.

Re: Irony

There was an interview with one of Trumps accountants. He said that Frederick Trump's tax records were meticulous. Frederick could prove to the cent exactly how much tax he owed. Donald's were a mess. It would take a determined effort to prove anything one way or the other. Donald did not show any interest during the meeting, but Ivana had many questions.

Re: Well, I have been saying this for the last 5 days

Firstly the idea that the data was to big to fit in the office has been roundly thrashed multiple times be people with more experience with large databases than me. As an example, years ago I read that Google kept a fixed size record of 128KB per user because that is the maximum Intel DMA hardware could move in a single command and it kept things simple. 50million times 128KB is 6.4TB so two cost effective spinning disks, or one slightly overpriced one. That is OK for a backup, but for a live system you would want flash. You can get ten times that on a single PCIe card if you can justify the cost. Space is not an issue unless you have a ten minute high def compromising video on most of the 50 million.

Even if that is the case, the database is not the only evidence of interest. There could have been evidence that they collected the data legally for a specific purpose. There could have been correspondence showing what the data was used for, perhaps beyond what the three people who read the privacy policy expected. There could be records of sales data without restrictions on use.

Do you really believe the Cambridge Analytica executives were terrified of the ICO drinking their tea.

Re: costs less to fly the same weight on BFR than Falcon 9

Actually Musk's claim was that it will cost less to launch a BFR than a Falcon 1. The first time I heard that I thought "What the Falcon?", stepped the video back and listened to it again. Not lower launch cost per kilo to orbit, simply lower total launch cost. I did a quick web search for Falcon 1, chose the most expensive launch and added a bit for inflation. Musk is aiming for a BFR launch cost under $10M. Over the next few days I saw other comments showing people were just as shocked as I was. It was not a mistake. Elon did not forget to say "per kilo".

If things go according to plan, you can put your falcon 9 payload in a BFR, add a Tesla Semi full of batteries for ballast, save $50M on launch costs and you get the Tesla Semi (with cargo) back on Earth. BFR will not be short of missions.

Re: my mistake

"It was not winter. Shepherds were out in the fields with sheep. People were travelling. Where does it say it was winter?"

My mistake, I thought Jesus being born on the same day has Santa meant it had to be Winter. I forgot that allowing for continental drift Bethlehem used to be in New Zealand.