Posts by Displacement Activity

NFTR

I had the same problem a few years ago. I would just bite the bullet and learn C++; you can read 'Accelerated C++' in a couple of weeks. It's not pretty, but it more or less does the job. My 'research project' is now getting on for 100KLoC, and I'm pretty sure I would be shafted if I'd used a proper language instead.

My next few w/ends will be taken up with trying to move it all from MinGW to MSVC.. :(

@I don't get it

Yes, it was common knowledge. This Daily Fail article was written before he was released, and says precisely where he'll go if he gets bail:

http://www.dailymail.co.uk/news/article-1338832/WikiLeaks-Julian-Assange-asked-judge-bail-address-secret.html

@DanGoodwin: a bit more background might have prevented the irrelevant chatter below from the Black Helicopter brigade.

@AC@it wasn't me

> Assange/ Wikileaks acquired the data, but there is no evidence

> that they acquired it from Bradley Manning - which of course, is the point.

I don't think it is the point; james68 is, I think, right. If BM/whoever had leaked the data, of his own volition, to the Wall Street Journal, do you think the US govt would be pursuing the WSJ? If so, how could any newspaper possibly operate?

Assange is just a sideshow here (assuming, of course, that he didn't actively solicit the theft). If he didn't have his head lost up his own fundament none of us would have even heard of him.

Re: Re: Windows tax: not all it's cracked up to be?

I don't give a s**t what MS gets. The "Windows Tax" is what *I* pay; that's what I care about. My life is way too short to worry about what MS is making. It's not about what "Microsoft ends up getting"; it's about what *I* end up getting.

The whole point of my mail was that, for what appears to be about £20, I get the peace of mind of getting branded hardware, which I'd pay anyway. I also get Win7 Home Pro thrown in, should I want to use it.

Re-read the title. I said it's "not all it's cracked up to be".

@Grease Monkey

Speaking as someone who does (some of this) occasionally. First, "The NHS" is thousands of different organisations, as pointed out above. What works for a hospital in Hull isn't going to work in a GP surgery with a dozen employees.

I occasionally have to do searches on surgery (and national) systems for audit and reconciliation purposes (for both clinical care and financial reasons); this is only going to get (much) more common. If the system can't do that, it's useless. I need to bulk-export data and analyse it offline. I'm going to do that back at base, not in a small building crowded with secretaries, receptionists, and patients. I take the data on a USB stick. Emailing it would be dumb, but no-one has explicitly told me not to do it; they've got their own jobs to do, without worrying about me. Besides, some datasets can be several gig.

I can't do *any* useful analysis without an age and a sex, and a treatment code, for starters. Sometimes I need an NHS number, to correlate a hospital patient with one of our patients. Sometimes I need a postcode, to get deprivation data. Names are never necessary. National databases already have sensitive information stripped out. However, there's generally something that they've stripped out that would have made life a lot easier. Surgery systems don't so this; it would just add another level of cost and complexity, and would be dumb, given that we already have the raw data.

Most surgeries do this sort of thing; it will be impossible not to do it post-commissioning, assuming that commissioning actually happens.

The data governance people don't have a crystal ball, and can't predict our requirements in advance, and how we want to use the data. You can generally use a "clinical care" get-out-of-jail card to justify what you're doing, depending on how anal your local people are, and what spin you put on it. And you can have a long argument with data governance, and might eventually get a nod and a wink, with nothing on paper. Life is a lot more complicated than a large book of regulations.

Ok, data will be lost eventually. But, at the end of the day, a lost database that contains an NHS number and a HES (treatment) code is not the end of the world. There is no HES code for "lost cucumber". The really sensitive stuff, joking aside, is only peripherally a medical problem, and there's no way any of that will appear in any sort of a database (except maybe at social services). You would actually have to physically read the GP's notes for that. There are other ways that this sensitive information can be lost - SystmOne being a prime example - but it's not going to happen through database analysis.

P/E

P/E is share price relative to annual net income (profit), not revenue, so it's much worse than that.

re 'A long term prediction"

Nice analogy, but not quite. We (in hardware) have already been going through the multi-cut/throw-away argument for the last 20 years. It's one of the major drivers of the entire (but small) EDA industry. Multi-e-beam shouldn't change this, but the dumber companies will still screw up.

The background is that we already do short production runs using FPGAs. These are expensive (and slow, and insecure, in that most of them are relatively easy to pirate). There has been an ongoing and gradual process of cost-reduction over the years, which involves replacing FPGAs and custom logic with ASICs.

Attempts to replace FPGAs included e-beam, over 10 years ago, and structured ASICs (I've done one of these myself). These weren't successful, and we're still in a situation where a short product run (say, <1000 units) is still much more expensive (say, a factor of 4 - 10) than one that ships 10,000, primarily because of the cost of FPGAs. An ASIC conversion of an FPGA is also generally much faster (in most cases, at least 50%). It's all about (1) cost, (2) performance, and (3) copy-protection.

It's certainly true that when people first started using FPGAs they cut many times and threw away everything till they got the one working version, and a lot of people still do this. This is possible because most hardware bugs manifest themselves a lot more obviously than most software bugs. Anyone doing ASIC conversions, however, even "cheap" ones, puts a lot more effort into verification, because it's just too expensive to fail.

No frickin title

I'm saying that there are 2 ways to frame an innocent person: plant stuff on their computer (#1), or make it appear that they planted stuff on your computer (#2).

In this case, someone pleaded guilty as part of a plea bargain arrangement, so it appears to be #1 (I'd be very surprised if security was enabled, by the way; your average pedo is probably not up to cracking WEP/WPA). I don't know about you, but when I hear the words 'plea bargain' I don't get a nice warm feeling that justice has been done.

I'm concerned because (1) in general, if someone doesn't confess, how are you going to tell the difference between #1 and #2, and (2) turning off security, hiring PIs, and waiting for someone to walk in smells of entrapment.

Post your own message

I get a call on my work number about once a month from the same auto-scammer. A recording tells me that the govt has introduced a fantastic new scheme to eliminate all my debts, with no obligation, and all I have to do is press a key to speak to an operator. I normally press the key, and wait for the retard at the other end to say 'hello' a few times, at which point they hang up. When this happened last week, though, I played along, and spoke to the guy at the other end for 10 minutes. I added a few decades to my age, and made up a lot of debts, as he got more and more excited. He gave me his company name, and then passed me on to his supervisor, who gave me a different company name. There was no government scheme, of course, but at the end of it I'd agreed to let this company contact all my fictitious creditors and reschedule all my fictitious debts, at a cost of only £100/month. He then sent me all the papers to sign, which I got a couple of days later.

A total waste of my time, of course. I don't think he's even done anything illegal, so there's nothing I can do with the papers.

@AC

Curious. This is an article about the reasons for failure in IT projects, and it touches only peripherally on the single most important problem. Your own very confused reply highlights precisely that problem.

Rule #1, IMHO, is to make sure that there actually is a *need* for whatever you intend to create. In other words, make sure there's a real problem, propose a solution, carry out a cost-benefit analysis. JS might argue that her second point ("...concerns the support and commitment of the different parties involved across the business, especially senior management") covers the first part of this, but GPs and nurses aren't part of the "business"; GPs, at least, are independent contractors who are nothing more than the proposed users of this system.

You (AC) state that there was a need:

"There is (and was) a strong need to allow greater access to peoples medical records. Put shortly IT WILL SAVE LIVES"

but you don't back this up. What breathtaking arrogance. Here's a simple sum: the annual primary care budget in the UK is about £125/head (that's what your GP surgery gets for *you*). This costs about £6.9B/year. The NPfIT budget was £12B. In other words, NPfIT consumed, or was intended to consume, the entire primary care budget for about 1.7 years. How many lives was NPfIT supposed to save? You've got no idea, have you? It's not your fault, of course - no-one has any idea, because they didn't bother to do the sums before committing to spending our £12B. If you ever do work out the answer, then perhaps you could also let me know how many lives would be saved by increasing that £125 budget to, say, £145.

And here's another rule. When your project crashes and burns, don't start talking bollocks at the post-mortem. Try to find out what actually did go wrong (there wasn't actually a "need", for example), or you'll have wasted all that taxpayer's money for nothing. In this case, I'm afraid that:

"However, there was a strong resistance from GPs in general, not because of privacy concerns, or about a perceived lack of need (indeed the SCR is replacing a chunk of the really useful functionality), but because GPs perceived the data as *theirs*"

qualifies as the bollocks. Back it up. I know a fair number of GPs and, without exception, their objection was *cost*.

No fricking title

> the pocket calculators, for which, I think, Sir Clive wrote some of the trigonometrical

> algorithms himself.

Whoa - didn't know that. My Sinclair Scientific took 14 seconds to work out the sine of 90 degrees. It's still down in the basement somewhere.

I used to work with Clive, on King's Parade - nice guy, good luck to him. I suspect that I'd recognise some of the names here, if any of us actually had names...

@AdamWill

1 - not much room here for technical arguments, but if you want an enterprise-level hypervisor the options are Vmware and Xen: Citrix Xenserver, VirtualIron, Sun SVM, Oracle Virtual Machine and Amazon EC2 are all Xen. No KVM.

2 - Not sure what you mean about the money. RH paid $107M for Qumranet (ie. KVM), then started dropping Xen. Coincidence? I don't think so. The $107M is doing the talking, and it says that RH wants to move into commercial virtualisation, and it wants to differentiate itself through KVM, and not by simply using the more mature (and GPL'ed) Xen.

3 - So, how many of the people who are paid to work on virtualisation at RH actually came from Qumranet? And how many of them run datacentres, as opposed to just developing the product?

4 - the Dom0 support in Fedora has never had problems (no more than anything else, anyway). That's why I (and lots of other people) are still on F8. The issue is that Fedora *unmerged* Dom0 support in F9. This was crazy - why? You don't need to be a conspiracy theorist to see RH's commercial interests behind this. Anyone using Fedora post-8 now relies on third party RPMs.

The answer for me personally, in the long term, is probably SLES. Others will hang on till full pvops integration upstream (Fedora 16? RH 10?), at which point you'll have no option other than to support Xen again, whatever RH's long-term roadmap may be.

@PatrickE

It's been a few years since I worked on phones, but the long and short of it is that handsets have a "top speed" below which they won't function. You get problems with Doppler shift, the time required to handover between base-stations, and power loop control. I forget the numbers, but GSM 900 probably won't work above 250kph, GSM 1800 about 135kph, and 3G even lower.

@What a convenient problem for TMO!

> This looks like a radio network issue, not an IP issue. How on earth would traffic

> management prevent a phone from switching its radio on and off?

As suggested above somewhere, the stack on the phone itself could detect this problem and deal with it, with no basestation intervention. Seems like a good idea to me.

Base stations already have the ability to order handsets to reduce xmit power, to prevent more-local handsets from shouting down less-local handsets. It would presumably not be too difficult to use this feature to cut off offending handsets.

Re: same here...

If it's public money, you should be telling us who it was.

errr....

...we can get more information from David Icke's website?

Title

There was a lot of packet-switched comms work in the early 60's, prior to DARPA and the ARPANET. The first UK packet-switched computer network dates from, I think, 1970, one year after ARPANET. There'd still be an internet without DARPA; we just wouldn't call it the "Internet". There'd still be a web without CERN; we just wouldn't call it the WWW.

The real problem with this sort of misguided legislation is that it will inevitably lead to fragmentation of the DNS as different countries try to take on local control. Most of the physical root servers are outside the US, and it's hard to see why any other country would want to play this game.

No frickin title

"This leads me to believe that it is perhaps the x86-32/x86-64 architecture that is at fault, at some lower level, for not properly securing access to 32-bit facilities provided by 64-bit processors".

I have a different suspicion, having just written (with some difficulty) a device driver that runs on i686("x86_32")/PAE. It turns out that, on x86, a significant portion of the driver core was written specifically on the x86_64 branch, completely separately from mainline x86. It wasn't back-ported, and x86_32 has significant problems, particularly if you're on PAE. This is made much worse by the fact that the 'documentation' and the relevant book are several years old, and talk extensively about features which the reader assumes will exist on any current kernel, but which in fact didn't work on the most common port at the time, and still don't.

In short, I'd be surprised if the processor is the problem here. Linux is my preferred OS, but it's obvious that there are significant management failings in the kernel development.

Title

As the Man From Apple says:

"The patent application in question does not claim as inventive the pictured user interface nor the general concept of an integrated travel services application".

This is presumably some sort of method and process application, and is seeking to protect only some combination of stuff running on an iSomething. The only people who need to worry about this are those who want to write apps to do the same thing on an iPants.

As to whether this is inventive, YMMV. Remember, the US is the country that has granted 5 (US) patents on perpetual motion machines, and on all elephant-shaped buildings. True.

@The Other Steve

Err... no. My wife's a medic, and frequently uses "Mrs" and "Dr" interchangeably, and has never had a problem. The billing cube/code monkeys aren't quite that dumb.

Not so sure about the "jealous and ignorant f***tards" either. In my industry, the PhD's joined 3 years late and had a lot of catching up to do. I don't recall ever meeting a PhD who insisted on being called "Doctor"; most of them would have died of shame first. Times may have changed, of course.

Post your own message

>The quality of IT support staff I want to take on when I and other GPs' organisations

> take over the useful* functions the PCT currently provides

There's no budget for that; good luck. As far as I can make out, there's no budget for anything. It's curious that the white paper suggests that some PCT staff could be redeployed to consortia. Presumably they're expecting them to do this on a voluntary basis.