Posts by John Riddoch 587 publicly visible posts • joined 12 Jan 2009
Yup, that's the point - Wipro assume they've been TUPE'd and therefore not their problem. Northgate assume they haven't been TUPE'd and are Wipro's problem. Either Wipro or Northgate should have continued their employment or given them redundancy terms/notice. Looking at the tribunal decision, they've decided that they were TUPE'd and therefore Northgate's responsibility to employ or offer redundancy terms to.
Oh, smoke alarms... first time I had one in my flat, it took me weeks to track down because it was beeping hours apart and it was only because it finally beeped while I was walking under it I twigged where it was coming from. New battery, problem solved...
Once you've had that happen, you at least learn that "random beep = new battery in smoke alarm", but the first time is a bugger.
Hard to tell how the Taliban will rule; they're making relatively progressive noises at the moment, but that could rapidly change once the world's attention span has moved onto something else. In any case, don't assume a close link with China. China has a chequered recent history with Muslims (Uighurs) and the more hardline sections of the Taliban might not want to be seen to be relying too closely on a non-muslim country.
Hopefully, Afghanistan can stabilise and result in a safe, prosperous country for its civilians and neighbours, but I'm not holding my breath. Too many political forces pushing and pulling in different directions.
IE also won because Netscape 4.x tried to do too much and did most of it badly. Netscape browser was decent, but they tried to throw email, usenet and $DEITY knows what functionality at it, clogging it up and slowing down the core function of web browsing. IE5 was about the time that it was actually a better experience than Netscape was and that killed it off. While ActiveX had its problems, so did Java. They both sucked, just in different ways.
As Partovi says, it then stagnated and didn't really improve until Firefox and Chrome began to be challenge them. With Edge now using the Chromium engine, they've basically admitted defeat.
It's pretty much about using the same toolset for automating stuff on-prem in the same way as you do for in the cloud provider's DC. And paying the cloud provider for the privilege. You'll save a bit on on-prem licensing (like VMWare) which softens the pain a bit, I guess.
It's not completely pointless, but it's a niche for when you need stuff on-prem (e.g. for latency) but don't want to have to manage VMWare and the other bits of the stack.
I seem to recall that part of the Itanium design was that the compiler would optimise the code rather than the CPU trying to figure it out at runtime, with all the out of order execution, branch prediction etc silicon that makes other CPUs more performant. To me, at the time, that sounded like a good idea; spend extra time at compile time to get efficient code rather than letting the silicon figure it out. In retrospect, it should have been obvious that there were several flaws here:
- It requires a good compiler; they were lacking at the time.
- It requires proper coding to maximise potential of the chip. Comment in the article highlights it was difficult to code for.
- Programmer time is expensive, more so than faster chips from Sun, IBM or even x86.
- CPU designs move over time; optimise for CPU x and CPU y will probably not work quite the same way.
Some good ideas in the chip, but the market moved before it was ready and it was practically obsolete by the time it shipped. AMD showed that you could have 64 bit CPUs which would still run legacy 32 bit code, so you didn't need to ditch everything to get 64 bit capabilities.
At least part of the rush to open up is allowing the NHS to be swamped in Summer rather than Winter when it's likely to be dealing with the usual seasonal flu & other bugs as well. I can see the logic of that, but it's going to be very, very hard for the ICU beds very soon, I fear.
In any case, their wishy-washy advice on masks is going to cause problems. Not only in terms of spreading COVID but also for the poor sods in retail/hospitality/transport who have to try and enforce it in their shops or whatever in the face of those who refuse to wear them.
No-one reads them unless they're bored, because they're too bloody long and lawyers often get paid by the word. There's plenty of fodder if you search for "how long to read eula" including the depressing statistic that it would take 76 work days a year to read all the privacy policies. Then they'll change them to give you another chance to read (or ignore...) them and click that you agree.
Almost everything he's done since November has had a side angle of raising money. The fund raising for "legal fees" to challenge the "election fraud" mostly went to RNC/DJT before any left over cash went to lawyer fees. Which probably haven't even been paid.
Companies like Fastly don't just do web hosting, there's a whole bunch of other stuff going on:
- DoS protection
- DDoS protection
- WAF filtering
- Content delivery/caching all round the globe
Can you do all of that for the same price as these companies and scale when some script kiddie takes a dislike to your website and tries to DDoS it? For MOST companies, the answer is no and the vast majority of them couldn't recover within an hour if it failed.
Yes, a lot of websites fell over at the same time, but I'd bet if they didn't use Fastly or another CDN company, they'd have fallen over repeatedly over the last year or two for other issues, it just wouldn't have been as visible.
It came with a CD, so chances are they bought the book and CD a while back and imported the images into their "textures library" without due diligence and have been using them since.
As you say, some of those could be coincidental, but the number of images seemingly replicated shows they've been using a lot of them. It's almost certain she'd win in court, but Capcom lawyers should be advising them to agree a settlement amount, probably along with an agreement to use all images in the future.
I think that was generally the whole point of it - a decent database/application platform with email tagged on as an afterthought, probably as part of Zawinski's Law - "Every program attempts to expand until it can read mail. Those programs which cannot so expand are replaced by ones which can."
There hasn't been any new Sun hardware for a few years now, it's all Oracle branded. However, there's likely still a market for old Sun kit on the 2nd hand market (which Wildfire seems to specialise in).
In terms of the lawsuit, it seems the gist is "we sold these servers in location X and you're not allowed to sell them on in location Y" which seems rather fussy. The grey market is generally used to get new kit slightly cheaper because companies priced kit differently in different regions to maximise their profit margins, I don't see the point in complaining about kit sold 10 years ago now being refurbished and sold somewhere else. Gut feel is that these companies have done something else Oracle don't like but they can't sue for that, so they're causing them hassles in other ways to get them to stop the thing they don't like as this feels like a shaky lawsuit.
I'm really hoping the standards makers have thought of this and made sure that chargers can't supply too much and devices won't accept too much unless they detect the appropriate conditions at all 3 points (charger, cable, device). Not easy to do while maintaining compatibility, but I know from experience that fast charging doesn't work with any old cable, so there's something there already.
Not all chargers or cables are equal and navigating the compatibility between them all is a pain.
Yes and no... software bugs are more widespread and usually easier to exploit, but they're also considerably easier to patch over. Hardware bugs can be difficult to impossible to work around; I don't think there's a comprehensive set of fixes for all the SPECTRE & MELTDOWN issues yet, at least not without obliterating CPU performance.
I think you're reading it wrong, I think - It's a combination of:
1. Old NCC Group internal training materials and content and
2. content that has either been incorrectly attributed to NCC Group or which is unconnected to NCC Group
What's probably more important is how much of (1) there is and how "old" it is (e.g. "we stopped using it when we got caught, so it's now counted as non-current"). The second part is likely a diversion - if they could find one page which wasn't theirs, they could accurately say the above statement, even if the vast majority of the documentation was theirs.
You've obviously not had the pain of "select the picture where the dice total X" so you have to peer closely at the dots on dice then add them up for the 4 or 6 image and inevitably it's the last image you find the correct answer on. It's not difficult per se, it's just time consuming and bloody annoying. Especially when the dumb thing asks you to do it 3 times or more because some algorithm has determined you're higher risk or whatever.
The defenders of this practice will say that the cheque from AWS UK to AWS America (although in reality, it's more likely something in Bahamas or another low tax country) is to cover the intellectual property/rights/stuff that the mothership does to make the product it sells in the UK worth buying.
Btrfs/ZFS can both do that pretty simply using snapshots; you'd just have to script a rotation of them. Alternatively, using cp -al/rsync there are ways to take snapshots within a filesystem, just make sure you don't have access to the backups from other systems on the network (i.e. don't share via NFS/SMB). I have a USB HD I take snapshots to in this way, means I have versioned backups.
OneDrive is "cloud". Windows is legacy desktop OS. I'd guess MS had a choice:
- continue to support OneDrive on Windows 7, keeping them in their cloud
- try to force the issue of getting them to upgrade to Windows 10, potentially losing a few OneDrive users/customers.
They've obviously decided on the former, prioritizing cloud over the desktop OS business.
Just like Oracle and IBM with their clouds. If you want to buy any software stack from them, they'll push you to use their cloud offering rather than on-premises because it'll be cheaper/easier/better/whatever.
Google and AWS have a different tack, since they don't sell on-premises software stacks, so the sales pitches are different.
You could say the same about demanding ID to create a social media account as is being put forward in various places, mainly Australia. By forcing ID, you kill off things like the Myanmar protests, because anyone speaking against the coup will be rounded up by the military.
We were never going to get the best deal with Trump (or any other US president). Unless their negotiators were stupid (and I doubt that), they would know that Boris et al would be so desperate for a deal with America to trumpet as a success to the press they'd accept almost anything.
Taking back control indeed.
The mention of "exchanging text messages" in the article implies short payloads and that may have made it easier for them to crack the encryption too. I'm pretty sure that short payloads are significantly easier to crack if you know the encryption and there's often "padding" put into short messages in real encryption to make it harder to crack.
So, yeah, poorly implemented crypto is a likely method of interception. I wonder how much they may have to reveal in the subsequent court cases on their methods?
I recall the adverts for Ariel/Persil (washing powder for clothes) in the UK in the 80s where they showed how much cleaner "new" Ariel was vs "old" Ariel. I think it was Jasper Carrot who opined that the tag for the version from 10 years previous must have been "cleans some of your clothes some of the time".
It's all very well saying "law enforcement" are the only ones who should get access (the old "if you've done nothing wrong" argument), but a corrupt law enforcement or regime WILL abuse that right. During the "Arab Spring" uprising, Western governments were very happy that the protestors were using secure messaging the authorities couldn't intercept/hack.
The same argument HAS to be used whenever anyone suggests hackable crypto. If the UK/US cops can get in, why shouldn't the Chinese/Russian/Myanmar/etc police and government be allowed to use the same back doors?
The core idea of replacing the serial init scripts with a parallel boot process is fine, it makes server booting faster and replicates SMF on Solaris in that regard.
Unfortunately, just as SMF absorbed a bunch of configuration into its clutches (e.g. DNS changes via svccfg rather than simply editting /etc/resolv.conf), systemd is doing the same. I'm not keen on so much being done by the init process in Linux - it's too critical and important to have bugs in.
They could well be that far behind. Smartphones don't have to go through a highly regulated sign off procedure with multiple countries' authorities before being allowed to be used. Remember aircraft have a lifespan of decades and some of those solutions are relatively new. While replacement/upgrade of the GNSS system in a plane is technically fairly simple (I'd assume), the red tape surrounding it would be much harder.
Just as a warning for anyone pondering a cheeky bid - the Rest and Be Thankful road (A83) is notorious for landslides and is often blocked for weeks at a time. Frankly, the main reason I've heard of it is because of the latest news story telling us the road is closed. Again.
For slightly different reasons, I can tell you the A939 runs from Cockbridge to Tomintoul. Reason being, it was always the first road closed by snow in winter time and if any road was shut, the A939 was too. This was from when breakfast TV was on in the morning when I was getting ready for school and the road closures would flash up at the bottom of the screen.
The term "grandfathered" historically refers to racist laws designed to prevent blacks voting but allowing whites to still vote. See https://www.npr.org/sections/codeswitch/2013/10/21/239081586/the-racial-history-of-the-grandfather-clause for a more in depth history of where it comes from.
To be fair, most people don't know that part of its history and I only found out in the last few months.
When I had to replace a headlight bulb in my car, the manual had a nice easy looking procedure. Plenty of space to get in behind the unit, take out the bulb holder and put in a new one. Reality? About an inch clearance for me to get my fat fingers into to release the bulb housing and be able to replace the bulb. I think replacing the fan belt (when it started squeaking) was actually less stressful than that process and that was a PITA for an amateur.
I sat my French exams in 89 & 90 with an English-French dictionary to hand. Theory was that if you were bad enough to need the dictionary too much, you'd lose time looking up words to be able to finish the exam in time. You needed the base vocabulary and could look up the odd word which doesn't come up often.
I've also done an IT technical interview where I could use Google as much as I wanted; was stuff like "configure Tomcat to do XXX" or "this MySQL DB isn't working, go fix it", so it was useful for looking up the obscure error messages or "how to" bits for things I hadn't done before on Tomcat.
I've also had to sit certification exams where some of the specific knowledge you need is obscure and not required very often, so it seemed odd to include it in such an exam. I could have found the answers quickly in the documentation, but that wasn't allowed.
Our kids have an O365 license, but that's mainly to help out with home learning with all the remote teaching going on at the moment. They were also using it for homework submission for the few months they were back at school, to avoid having to hand over (potentially) germ covered submissions.
We'll see if it's kept going once things get back to some kind of normal.
I don't think Vi (well, originally ex) was designed to specialize for coding. It was written in a time of scarce, expensive compute resources and dumb terminals wired by RS-232 cables. Minimising keypresses reduced the amount of data being transferred over the wire and being managed by the operating system so it became efficient as a by product. Mice weren't available much, if at all.
It's a steep learning curve and frankly I only got good at it because it was the only editor guaranteed to be available on Solaris, AIX and HP-UX on a base install. I still use it as my default editor on Linux despite other options being available out of familiarity and the fact I can do a bunch of common tasks on config files quickly with it (quicker than I could with e.g. notepad). Also, it's quicker for me to not have to switch between mouse & keyboard, just keeping my hands over the keyboard.
If you prefer another editor, great. For folks like myself who can be efficient with vi(m) and are familiar with it, it could well be faster than using a GUI based editor.
I get the joke, but realistically, this is a niche product which is going to see very little take up. The folks who really want one will buy it anyway and it needs a decent mark up to cover their development costs (which have to be spread across fewer units). The price is about right for what it is. They'll probably still lose money overall, hoping the development costs and learnings can be applied to some other future product they can make more money on.
It's not something I can see any use for, but that means I'm not part of their target market and I'm fine with that.
If I'm searching for a term, say "widget", it's probably trying to find widget.com, widget.org, widget.co.uk, widget.net and a whole bunch of other domains. Yes, those queries will go to my "local" DNS server (which could be a corporate one or my ISP). Unless they've cached that information already, they need to look to the root servers to find the answer, hence it hits the root servers eventually. They should cache that for future requests by the next person searching for those domains reducing future traffic, though.
It's perfectly possible to get locked into a single cloud vendor stack if you don't plan to avoid it. Using, for example, Amazon RDS, Lambda, AWS ID services extensively and it's non-trivial to move to Google, MS or Oracle cloud without a lot of work. If all you're doing is apps on virtual servers & storage, it becomes much easier once you plan for the vagaries of each cloud platform.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
