Posts by Ken Hagan

Re: Value for money??

"I had a friend who advertised a new product through fb with particular demographics specified. They got increased sales from the campaign and given that it was pretty much their only marketing, its reasonably safe to say it worked."

It says nothing of the sort. You say "advertised a new product" which strongly implies that you changed the bleeding product on offer to punters! It is hard to imagine a more significant variable to change prior to hunting for a change in sales figures.

Re: I'm ignorant! But I'm commenting anyway!

"the best they could find were these two boyos ?"

Sadly, the US does not have a monopoly on crap politicians.

All the more perverse when you consider that the maths in the diagram is secondary school level, at most.

Does that mean that Oregon schools are festering holes of criminality, pouring illegal ideas into the minds of the young, at the taxpayers expense?

I don't see any Scottish motorways and it isn't clear whether the "M4" includes the Welsh bit. Looks like a survey of English motorways to me, with the stunning conclusion that the motorway with most tarmac in a hilly part of the country has the worst coverage.

"The Sun and The Daily Fail can then get off their high horse."

You're assuming that the Fun and the Fail don't get caught in the cross-fire. I could imagine a law like this being used as a stick to beat almost any large web-site with a lot of end-user-content in its comment pages.

"Please let me know where it says that In the article?!"

Well we can start with the title: "blow ... off the net".

My point is that if your choice of ISP results in you using a modem that gets DOS-ed of the net entirely then it does not matter how fast that ISP promised to deliver bits. You are getting 0 Mb/s and any ISP that offers a better modem will deliver more bandwidth.

Re: ??

"the highest linpack score given a power budget"

Ah, thanks. That does clarify things somewhat. However, isn't LINPACK embarrassingly parallel? If so, the competition would be mostly about deferring your purchasing decisions until the last minute (to get the shiniest new things in your box), so what's the point?

Re: More money for CEOs guarantees poorer performance.

That doesn't surprise me. If we assume that execs have no effect on company performance at all, but their pay rises and falls with performance but plus a several-year time-lag, you'll get exactly that effect and none of it means anything.

Re: Quarantined *signed* files?

There's a system for compromised certificates and whilst I've heard stories that CAs have been tricked into issuing certificates that say "Microsoft" on the front, I haven't heard stories of rogue certificates that have been counter-signed by Microsoft's own root certs. And unless you (or an AV vendor) can find an instance where this happened, I maintain that the presence of such a signature proves beyond reasonable doubt that the file in question is not malware. Quarantining it is just reckless and proves that the AV vendor doesn't care about trashing your system.

Re: Microsucks is comical

They'll probably exist, since they have a very large cash pile to run down. (Hey, if SCO can keep coming back years after the money ran out, MS will probably outlive all of us!)

What they won't have is any real products, just the fat-client-as-a-service thing. And since business likes to amortise hardware over more than a few months, they will soon find that they have a fat-client-for-thinning-customers thing.

Re: Ask my permission

"I cannot see why websites or apps or any other logic running on my device should be able to do *anything* without explicitly asking my permission."

I think the bottom line is that end-users are too stupid to know whether to grant that permission or not.

The fundamental flaw with ActiveX, as originally envisaged, was that you had no control over the code that was running on your machine. Microsoft addressed that by adding a "Do you want this to run?" question and using code signing as a means of helping to answer it. However, in practice most users had no way of knowing whether it was trustworthy or not and simply said "Yes" because otherwise the web-site didn't work.

Java tried to build a sandbox so that there was no need to ask the question. That approach was limited by the fact that a sandbox good enough to keep you safe was also too good to let exciting things happen, so inevitably there came a basket of special permissions that you could grant and web-sites didn't work unless you granted lots of them. Modern-day Android users face the same problem and answer it with the same "Meh, whatever!" response. (Sandboxes also appear to face quality of implementation problems, which is odd because an OS isolates processes in the same way and yet privilege escalation bugs in OSes are quite rate compared to sandbox breakouts.)

Javascript appears to have begun life in a sandbox and is now desparately trying to shake that off to become more ActiveX-like. Quite why programmers are pushing for this is a mystery to me. Of all people, you'd have thought that they would be able to understand the risks and remember the history.

Meanwhile, traditional desktop apps are relatively safe because they tend to come from either people you know and trust or people who have a commercial reputation to lose if they mis-behave. Neither of those constraints applies to "crap slapped on a web page by a third-party ad-slinger".

Quick poll: Do people in business meetings now get asked to turn off their phones, or has it not yet penetrated mainstream consciousness that these devices are constantly streaming audio back to a server somewhere in the US, where it is subject to the attentions of a TLA that is constitutionally required to further the interests of that country.

You mean "scripts" rather than "languages" but, yes, I suspect that this is how the issue will be resolved.

I believe there is some opposition to this on the grounds that several thousand people have legitimate registrations that would be classified as "mixed" under your rules and would therefore be penalised despite doing nothing wrong. Yeah, sad, but sometimes a few shits spoil things for everyone else and I think this is one of those times.

Re: Proper procedure?

"They should [test them that way], though"

I beg to differ. I can't see the sense in asking for a patch regime that is literally untestable!

Two to the power N is a very large multiplier on your test matrix, even for fairly small N. If MS actually allowed users to install some patches but not others, and the same again next month, and the month after, the explosion gets to unfeasibly large N probably in the first month and certainly quite soon thereafter, no matter how large your test farm.

I suspect that's why MS no longer allow you to do that. I suspect also that patches affecting the same component are always cumulative. (I don't think the roll-up bundles are, but they may be "chained" so that you have to install all previous ones before any new one. If not, they probably will be soon.) Under this regime, N months of patches gives you at most N possible configurations and each month there is only 1 new one that they have to test because they did all the others in previous months.

"1" scales to large N in a way that 2**N doesn't.

I would also note that most Linux distros work this way by default and even if most (all?) offer ways of pinning packages to a particular version, they don't promise that an arbitrary combination of pins actually works.

Re: But 72 layer though...

Not sure about that. A 9:8 ratio of physical capacity to advertised capacity could just be the spare pages that they keep to replace worn out ones.

Re: Who could possibly design a good fake news algorithm?

To give but one example, when the article boldly asserts:

"Detecting false advertisements, bullies, and bots - all of these can be done with machine learning. It can even be applied to a politician's tweets - to find out if they’ve been fibbing about where they’ve been, and when."

...then that's fake news. In most cases, even human beings can't agree on whether "fake news" stories are fake or not. It is clearly beyond current "machine learning" to do so. Can we just put all this AI/ML crap back in the bin from where it was pulled by a bored journalist a year or so ago?

Re: Just a matter of timing

"This won't work, for the same reason that we don't get giant insects - as you increase all the dimensions the corresponding cross sections are squared but the corresponding volumes, and therefore mass, are cubed."

But the power-to-weight ratio of a battery stays the same, so it rises with the cube of the linear size of the battery.

Also, the size of insects is limited by the way they breathe, not by their mass. During the Carboniferous period when oxygen levels were much higher (35%), flying insects (and many other forms of life) were much larger.

Also, also, the effectiveness of wings increases with area. This increase is slower than mass, but you can offset that to some extent by changing the proportions of the plane.

An electric plane doesn't need to be as good as a kerosine one. It just needs to find a niche where it is more competitive than anything else.

Re: BS 5216C:2005 Proof correction marks (Pack of 20)

'Are they big enough to write "this document was written by an idiot with delusions of adequacy"?'

These days, yes, since there's a Unicode character for that: http://emojipedia.org/pile-of-poo/.

Re: Ubuntu Touch

"download the code and continue development!"

Ummm, so if I am daft enough to want a bit of privacy on my phone, I need to become a system's level programmer, an application's programmer, build my own phone, connect it to my own mobile network ...

The point about Ubuntu Phone was that Shuttleworth was one of the few people on this planet who actually had enough money to pay for all those people to come together. Now he has lost interest, apparently because his preferred phone UI is (like every other phone UI) crap on a full-size screen. I don't see the logic myself, but apparently it makes perfect sense to both Mr S and to Microsoft, so it must be true (despite Apple proving every day that it makes far *more* sense to use different UIs in each case).

Re: Wrong

"... the engineering that goes in to a modern car is immense, but fundamentally it's not new."

Even more fundamentally, it is nothing to do with driving. The self-driving cars that people are working on at the moment have three parts: a car, dozens of sensors, and some software.

The first is a commodity. All the tech firms who are doing research in this area have simply bought a car and modified it. As and when they perfect their algorithms, they will either license the technology (perhaps just to the big auto manufacturers but perhaps more widely) or they will simply buy cars from the cheapest manufacturer and modify them. It's not like Google or Apple don't have cash piles that would help them get this sort of business going, or like there would be a shortage of investors willing to help.

The second is also a commodity but not a commodity currently made by car manufacturers. The likes of BMW are no better placed than the likes of Google in sourcing this stuff or fitting it to a vehicle.

The third is the subject of research and It Seems Quite Likely To Me that the companies that are best placed in *this* race are ones that specialise in doing clever things with algorithms. The last I heard, the auto industry's IT departments were trying to fiddle their emissions results whereas Google and Apple were taking on problems like machine translation, speech recognition and machine vision.

So, yeah, it's hard to know where the smart people are going to put their money.

Re: The "Serious Fraud Office"?

Off the top of my head ... the SFO was set up as a result of a growing belief that existing ways of prosecuting fraud only worked if the defendant lacked the resources to just throw lawyers at the problem until it went away. The "serious" was therefore a reference to the Establishment's belief that "untouchable" defendants are a Bad Thing. As such, it refers less to the actual fraud and more to the wealth of the person alleged to have committed it. I have no idea where the boundary lies. Perhaps this post will be sufficiently wrong that someone with a clue will jump in.

The fact that Private Eye calls it the Serious Farce Office would suggest that it hasn't been entirely successful, but the original idea was probably sound.

Re: ... start your vehicle and warm it up from inside the house on a cold day...

Once it has used up most of the oxygen, even a really efficient engine might start chucking out CO.

Re: Insert open-source licensing FUD

It is not the attempt to contact contributors that is alienating the community. It is the fact that the community will be ignored if that one attempt fails. Happily, bobajob12 puts it nicely in a post which is (at time of writing) directly below this one, so...

Re: Is discipline across government too weak?

Central Purchasing would not allow this kind of thing to exist ensure that the kickbacks went to different people.

Centralisation merely changes who is able to be corrupt. If you want to put a stop to it, I suggest a healthy dose of transparency, which is effective against all but the most brazen and which works on pretty much any scale.

Re: even worse...

If you re-read the fine article then you'll see that Dabbsy already covered that one...

"a new button you’ve never seen before labelled Delete File, Overwrite All Backups With Zeros and Hack Off My Scrotum With a Rusty Saw."

The phrasing is different, but the meaning is the same. Perhaps you are using a different locale.

Re: not-so-smart with pointers

"While that's better than nothing it's certainly not as good as proper intrinsic types that can be checked by the compiler, generate efficient code and meaningful errors during compilation."

If your compiler cannot generate efficient code from source then you need a new compiler. If your language doesn't let you state the constraints that apply for given types, you need a new language. If you need more meaningful errors than "look at this line -> here" then you need a new programmer.

The advantage of implementing this stuff *outside* the language is that any third party can add to the toolbox. You don't need to wait for your compiler vendor to implement the feature and if you are trying to write portable code then you don't need to wait for *every* compiler vendor to implement the feature.

Re: Rule #1 for the user-facing components development

"poor souls working [...] on ASP.NET web apps. This approach would guarantee nothing will ever be delivered."

You say that like it would be a bad thing.