Well there's also the US election result
The one where the dollar fell as traders sold and withdrew to their bunkers before the coming apocalypse...
“My own story would not have been possible but for the democratizing force of Microsoft technology reaching me where I was growing up,” CEO Satya Nadella told shareholders this week. But the price of that “democratizing force” is about to go up, with Britons uniquely singled out. Microsoft has reiterated to Azure customers …
They are simply adding in the expect crash in exchange rates in one go rather than having to come back for a second hike in March. The problem being that the doom and gloom is coming. There will be no single market access without freedom of movement and hefty payments into the EU social fund (approx 90% of what we are currently paying). As such the weirdo bellends in the Tory party will have screwed over our manufacturing base in the 1980s and our services sector in the 2020s.
>> "For new Office or Azure cloud customers in the UK, no exchange rate can justify any price rise at all".
You are aware that data centres, no matter where they are located, require quite a lot of gear not made in the UK and mostly priced in, or at least with prices based on, USD, yes...?
The price increase percentage being well-above the GBP price collapse is arguable, but it's probably just indicative of what Microsoft expects the future will bring.
Anyway, price increases from foreign service providers is a boon to local service providers, no? Think of it as an opportunity for domestic IT services!
This is what many people are missing in subjects such as this - Microsoft pay a lot of money up front to fit out their datacentres, and it costs them a lot of money in both engineers and replacement kit to keep it going.
They invested a lot of money up front to get this going, with a view to making back that investment with a profit over many years.
Therefore, price-setting is done based on forecasts of how much kit and talent is going to cost them over the next few years, whilst trying to make back on the initial investment as well.
They need to maintain a "war chest" for kit replacement, based on forecasts of how often kit will need replacing and how much that kit might cost. And they need to get back that investment such that over 5 to 10 years they make an overall profit.
Trying to pretend it is all about the current exchange rate and current costs is very dubious. And using such poor logic to bootstrap another argument about how angry we should be and how hard done by we are by our own poor decisions is astonishingly bad taste.
And just to caveat this, I HATE being put in a position where I need to defend or justify MS. But I hate sloppy logic, terrible critical thinking and poor foresight even more.
But said equipment and service management, excluding local employees and datacenter location costs are in USD, so the cost of purchasing services to be inline with these costs would be need to be in USD, so the exchange rate is relevant.
As was said in the article, if it was in USD then everyone would pay the same. Does the price in the USA change for people in the USA also? No, so the current exchange rates do matter..
"At the moment it looks as if they wand to do it in one year."
Do what in one year? Recoup an unknown amount of money that you and I have no idea of, at a rate which is unknown to us?
I think you've lost the thread, Ivan: You've made an assumption which isn't based in facts we are in possession of.
This is no different to when your bank changes your mortgage interest rates by more than the BOE base rate has shifted: They are (ostensibly) trying to hedge against risks and expenses so on that we aren't privy to, based on timelines we don't understand.
The 'over many years' is the main point here. At the moment it looks as if they wand to do it in one year.
They're just being practical. Normally they plan for returns over many years, but if they foresee that within a few years the UK customer base will have reverted to bartering leaves for acorns for food, and have moved on to a post-technological society (counting on their fingers), then they're going to get payback while they can. The only clouds in post-Brexit UK will be the ones that pour rain on our crops.
"If they foresee that within a few years the UK customer base will have reverted to bartering leaves for acorns for food"
With our lords and masters wanting to tamper with crypto for their own ends, nobody sane will want to take credit card orders from us. We'll be back to Postal Orders before we know it... but they've pretty much buggered up the Post Office in rural areas.
If you'll excuse me, I'm off to gather up leaves.
>The 'over many years' is the main point here. At the moment it looks as if they wand to do it in one year.
They're looking ahead 3 or 5 years and also factoring the 'hit' of last year or so based on getting their previous estimates of currency fluctuations badly wrong - MS analysts certainly weren't expecting a Brexit vote. This is no different from other US providers and suppliers. The Pound will remain low, it will most likely drop to around parity next year.
If only for PR's sake, they should take a leaf out of Amazon's book and simply charge explicitly in Dollars + local tax.
"They invested a lot of money up front to get this going, with a view to making back that investment with a profit over many years.
The 'over many years' is the main point here. At the moment it looks as if they wand to do it in one year."
I seem to recall that Microsoft's MO is to have the minimum number of price changes possible. At every price change they try to predict future trends and price accordingly.
If I'm recalling correctly, I picked that up from a Reg article on enterprise VLAs some time last decade.
Anyway, price increases from foreign service providers is a boon to local service providers, no? Think of it as an opportunity for domestic IT services!
Yep. But of course they'll have to buy all their kit from abroad (unless the Govt plan to resurrect ICL?) at foreign prices in dollars (plus tariffs if they don't move fast).
And lets be honest, unless they're forced to, who will want to store their bits in the great May surveillance state?
<footnote>See how I managed to get in a dig at Brexit AND the Snoopers Charter in the same comment. I'm cooking on gas today!</footnote>
There is a corner of Fujitsu that is still to all intents and purpose the rump ICL, keeping the now-virtualised mainframes spinning and supporting half a century's worth of legacy application software. It's one of the bits that makes it a profit, so it would only take a change of nomenclature, branding and a lick of Burnt Tango paint and it could happen very easily.
"There is a corner of Fujitsu that is still to all intents and purpose the rump ICL, keeping the now-virtualised mainframes spinning and supporting half a century's worth of legacy application software."
There is indeed, I did a stint up in Stevenage about 10 years ago at the old ICL building when Fujitsu still owned it, now even Fujitsu have vacated the building.
>You are aware that data centres, no matter where they are located, require quite a lot of gear not made in the UK and mostly priced in, or at least with prices based on, USD, yes...?
I think the point is that the DC's are already up and running. Incremental costs do not justify the price hike across the board. E.g. getting UK engineers to service the DC just got cheaper, UK electricity just got cheaper. Imported stuff is not the only cost so even passing on the full cost increase wouldn't give you the full currency difference. All the income from non-UK accounts using UK DC facilities just got far more profitable... and yet, price gouging.... because we can.
Welcome to the Cloud. We own your data, we own your infrastructure. We own your business.
We also know that if you had a choice, you'd probably have already gone to AWS. So we're guessing there's an inelastic demand curve in play. Since the global economy has gone down the toilet, we haven't really added anything new and beneficial and we can't sell more stuff, we're going to have to hike prices to maintain our margins. That's not a long-run strategy, but I'm going to make sure I get my bonus.
If you own your own IT, you can sweat the assets in the hard times. The whole point of XaaS is to stop you doing that. Did you really build your business profitability on the basis of favourable exchange rates, did you? The price hike is bad enough, what happens if your supplier goes out of business? How long will it take you to find another supplier? Ah, you've tied your business to a single supplier with no hope of an alternative due to IP laws protecting the APIs? We're not just talking "unsupported" here, we're talking "gone." What if your supplier is Lehman Bros?
IT vendors have been vertically integrating because you can't trust third parties with your business. IT users need to learn the same lesson.
TL/DR; The economy is shot and the Emperor of Eternal Increases in Profitability has no clothes. Cut-price goods and services are usually shoddy value.
So are the British being punished for exercising their political choice, and choosing to leave the European Union?
No. What you are missing is that at such sizes and scales prices (as they are set long term in volume contracts) reflect not only current supply, demand and cost. They also reflect what the seller sees as the supply, demand and cost across the median future of the pricing agreements.
On the positive side, Microsoft forecasters see the demand and supply remaining relatively healthy. On the negative side they see the pound undergoing further tanking. Is this forecast correct or not? Nobody knows. If the forecast is wrong Microsoft will lose money because people will move to other cloud suppliers (if they can). After all the customers who buy A LOT often have good forecasters of their own as well.
IMHO the forecast is probably better than a lot of forecasting by our politicos and civil service droids. End of the day, Diskworld comes to mind. It is the difference between "doctors" and Doughnut Jimmy. While Sat Nad is not known to break chairs, it is still probably not good career guidance for the forecast people to make him "very upset" (in the Mr Chrysoprase sense).
their business plan of how quickly they plan on hitting profit on their original investment:
Thanks for reminding me, this one is actually quite important.
Anyone who thinks that this metric has not changed as a result of BrExit needs to have their brain examined. In fact, the ROI periods used in calculation have shrunk and their cost contribution has gone up. This drives the overall price of goods and services which rely on capital investment (in this case - servers, datacenters, datacenter equipment).
This is not for any other reason, but because of uncertainty. From this point of view the economy and the prices for services and goods relying on long term investment would have been BETTER off if there was a cliff nasty cutoff with a given date. It is something you can plan for. Exit on date X with tariffs Y and restrictions Z - fine. That is something you can stick into a forecast and plan BEYOND that date.
The current woolly grandstanding by the 3 blind mice - not a chance. Any financial planning more than 2 years out are in the realm of non-scientific fiction. So whoever in Microsoft turned the ROI knob up a notch simply did their job.
I do believe Amazon and Cisco have it right as do some others (non-tech)... price it according the "home base in home base currency". If I buy, and I do, stuff from the UK, the suppliers I use the pound. From France, it's the euro. I pay what the exchange rate is in dollars. Very fair.
MS does seem to be taking advantage of things... and given the weasel wording of the release... hmm.. makes one wonder.
Pricing in a foreign currency lets companies exposed to sudden prices increases (and decreases, OK, but that's usually welcome). If you're going to spend a lot of money in a subscription like could services are (and use that for critical tasks), you may not want to be exposed to sudden big changes in its costs - otherwise you would need to plan for it also, just like in other sectors companies need to carefully manage supply prices.
MS are dealing with marketplace demands. As a major player, they're dealing with essentially commoditised services. They can't have highly reactive pricing, because most of their customers don't have reactive pricing for their customers; pricing in US dollars would effectively be instantly reactive pricing dependant on exchange rate movements only, and making smaller adjustments to pound price to account for exchange rates would be literally reactive pricing. That means dollar pricing would bugger up MS's clients' budgets very quickly indeed, which would be total disaster in international markets.
Microsoft sells only in dollars.
Market movements lead to deflation/devaluation in Xland.
Speculators then consider that the majority of companies in Xland are going to have a soaring cost-base, so consider all such companies a liability, further destabilising the economy of Xland.
Microsoft's policy of making large, infrequent corrections to account for exchange rate moves is actually pretty good for everyone.
It's very embarassing to raise prices every three months. Maybe they're just planning ahead for the next falls in the pound once the governments 'strategy' for Brexit is finally revealed (does running around like a headless chicken and pretending black is white count as a strategy?).
Pound fell 19%, US company puts up prices 22%. That's actually fair: note that (1/0.81) = 1.23
Of course what really matters is whether there is sufficient competition in the marketplace for cloudy office services.
The only real competitor is Google, and suddenly Google Apps looks a lot more attractive - as long as they don't follow suit.
anyone using Raid-0 for business critical data needs to be taken outside and shot, hung, drawn and quartered (IMHO) and then burnt at the stake if thay are still breathing.
As a minimum you need Raid-1 (Mirroring) not Raid-0 (Striping)
I've just had a 5-drive NAS spend 16 hours rebuilding itself after one 4TB HDD (WD) went Phut.
Thankfully it was configured as a Raid-6 array.
"RAID0 can be more reliable than RAID5" is an example of very dubious arguments. The basic points are:
In RAID-0 any drive error is losing you data, and typically a lost HDD means you have to wipe and restore the whole file system. Only upside is you *know* you are vulnerable so probable (I hope?) have a backup and restore plan that is regularly tested.
In RAID-5 you can tolerate one disk fault, be it a whole disk or reported bad sectors. Down side is folk over-estimate the independence of errors and the correct reporting of errors. If using any RAID system you really must do regular disk scrubs to make sure that the inevitable rebuild has a sporting chance of completing OK and not throwing up other errors.
However, if your data really matters than (a) you have a backup anyway, as RAID != Backup, and (b) you should be using something with double parity, at a minimum RAID-6 or better still ZFS RAID-Z2 since it has better write performance (more so with a SSD for the intent log) and additional checksums on the blocks so it can spot HDD lying or disk controller faults, etc. I think btrfs is planned to have a similar scheme (i.e. redundancy and extra checksums like ZFS) and some variants of GPFS (or whatever IBM call it now) has it, but on the payment of lots of extra money.
"In RAID-5 you can tolerate one disk fault"
Well, that's begging the question; the claim of the linked article is that for a certain, currently realistic, size of RAID5 array the likelihood of an unrecoverable read error during rebuild is as good as certain making the redundancy useless and meaning that a RAID0 array of equal capacity is more reliable as it has fewer drives.
It's not dubious, it's a perfectly logical argument, although I can't vouch for the actual figures used in the calculation.
It's not a great argument for the use of RAID0 and doesn't seem to be intended to be but it is a strong warning against the use of RAID5 in large arrays.
I thought it was interesting.
I found Microsoft were selling the Xbox One S (Minecraft edition) cheaper than anyone else last (black) Friday. £199 compared to the £229 everyone else seemed to be flogging them at. And they added an extra Forza game.
Maybe the pound had done a funny u-turn or something last week.
Biting the hand that feeds IT © 1998–2019