Why would there be any retries? You don't get a four day delay when you email an address with a typo and end up with John.SAmith@example.com instead of John.Smith@example.com now do you?
You just get the non-delivery report straight away.
150 publicly visible posts • joined 1 Feb 2011
I suppose at some point they will consider allowing Azure stack hci (the premium hypervisor they want to push as a replacement for plain old Hyper-V clustering) to use real authentication to onboard then, not device flow?
Nah... Why do that when you can just blame others?
It seems to me that it's all in weird custom form factors and connectors though. Your pokey little m.2 SSD outperforms SAS by a factor of three or more.
The server SSDs are suited only for the server grade kit they're plugged into and even the servers themselves are often custom sizes, or need liquid cooling.
Frankly it's a mess
The underlying premise is that we have and need more computers than ipv4 addresses.
If you use class D and E space there will be lots of systems you have to replace because they consider those IPs invalid.
So obviously you can't do that. And even then it's a stopgap.
If you make the addresses longer then that's just IPv6 with a different number of bits. There is no possible way for an ipv4 only system to communicate with anything other than ipv4 systems because there's literally nowhere in the packet header for the extra bits.
And even then how do your routers know where to send the packets? They all have to understand and expect whatever new thing you're proposing.
IPv4 is by definition and by implementation fact exactly 32 bits and you can't magic in more bits while still keeping it only 32.
It's always felt to me as if disgruntled should mean someone who has been deprived of their gruntles, and therefore that someone might instead be ungruntled, implying that they never had any gruntles with which to be gruntled in the first place.
Sure. And it's just coincidence that Google's own synchronisation app just happens to be the only one that won't be summarily broken by the change.
But hey choice is bad, just upload all your data to Google what could go wrong? No-one should have the right to avoid bundled services from the company that makes the phone OS right?
> My work isn't organised or indexed by date; it's organised and indexed by project; I suspect that's true of most other people, too.
Of course it is. The point of recall is to ask questions like, "what did I do on the x project last week" or, "what was the name of the library I researched for chicken sexing a few days ago.
You have to understand that most people struggle to remember what they're presently eating let alone what they did a week ago.
I know it's the slippery slope argument but I'm still predicting that six months will go by and you'll be able to get cloud copilot to "assist" and mine more before it becomes automatic opt out.
So somehow it's Microsoft's fault that a third party wrote a bad driver ...
You can argue you shouldn't get a stop error but you're ignoring thirty years of history, including third parties whinging that they were being locked out if Microsoft didn't allow them to write and install those crap drivers.
And by legitimate senders I mean things like mailboxes for your helpdesk platform for customers, your CRM etc. Just doing the maths for one app we have, it already has an extended-for-us limit of 2500 external recipients a day and we get to about 30% of the limits on a normal day and 70% on a busy day.
It's integrated with office 365 because that's where our other mailboxes are.
And now Microsoft wants more cash for the same function. We could send much more with on prem Exchange when we needed to, but cloud says no. We're not even a huge company - fewer than 20 staff.
And yes, we'll need to switch to ACS or similar. When the app supports that configuration, which it doesn't now and it isn't on the roadmap - oh yay more development cost because Microsoft wants more profits.
You said, "There is a better way and that's proven by the fact other OS don't have that same problem."
Unfortunately you seem to be unaware of the concepts of a UNIX/Linux/BSD/OSX kernel panic, original Mac "bomb" crash, which are all pretty much the same thing.
Something went wrong and continuing with error checks might corrupt things, so the safest thing to do is "Halt and catch fire". And it's a balancing act. Want stability? It'll cost more and/or be slower. Pick your poison.
There are a couple of different options I have found. AOOSTAR has some NUC-sized AMD units that have dual 2.5Gb networks plus wifi, three PCIe SSDs and up to probably 64GB of ram in the new one, and MinisForums put out the ms-01 recently with I think three SSD locations, dual 10Gb and dual 1Gb networks and a PCIe slot. That one supports 64GB but unofficially I've seen reports of 96 GB working.
I have a recent AOOStar unit and so far, so good.
What I haven't managed to find is a NUC / NUC+ sized box that will take 128GB or more.
Incompetence you say? All they did was create an internal API to unquestioningly and permanently delete things by number, and passed the wrong number to it. 800+ times. Because no-one noticed the wrong numbers were supplied.
Could happen to anyone!
Paris is looking at the list of IDs to be submitted to the API...
It does look like a great idea and is probably a good tool, but demands openSUSE for the server. So unless you're using it already, it's another Linux to learn and understand (while the base is the same the foibles matter a LOT).
Were it possible to deploy to Debian or Centos then maybe I'd consider it more thoughtfully.
I literally deployed Azure CDN today, brand new for a customer. The beauty of CDN plus storage account was low cost - I had a small static website that would have been literally cents per month. This change ratchets the costs up about 10000%. Couldn't you have waited a week till the project closed!? I would have avoided all the angst!
The problem is that people don't read them anyway. Perfect case in point would be an error message that says
"Please call the help desk on 555 5555 and say that the Rostrum app had error Bingo"
They call and what you're told is that
* I can't log on
* There's no error message
* Windows is broken
* It's definitely not the same Bingo problem in Rostrum that twenty people have reported.
How do you make that any clearer?
I'm not so sure it's entirely on the dev. Surely part of the check before releasing the name should have included checking simple things, perhaps like "whether the bloody thing had been updated in the past year". I recognise that there are statements about "fixing the process" but really, given there was evidence of continuous use, it should be more like the ICANN result where the original owner gets it back.
Just being deliberately obtuse perhaps, but this stance is consistent with the US stubborn belief that perfectly free markets are wonderful and business is the only thing worth supporting.
Allowing clawbacks for fraud opens businesses to such oppressive actions like having to deliver what's been sold lest the business have to refund.
I know it's not actually that had but it sure seems like it sometimes.
I also prefer Core where possible but it's no longer true (if indeed it ever was) that there are fewer updates and reboots. Microsoft shot that idea down the moment they moved from individual updates to the CU model. Core and Desktop receive the same packages, same size and they install in about the same time on each.
I suspect I'm neither the first nor only person to call it the Cumulative Update for New Technology model on the back of Windows historical designation.
For goodness sake. Windows 10 is more than five years old at this point. Windows 7 hasn't received security updates for a year.
You can pretend all you want that you're waiting to see what happens but reality says you've made the decision already to stay on old and familiar but increasingly insecure.
In my view you're the "XP is best" dinosaur from late Windows 7 timeframes.
I've considered the idea of solar + battery cars in the past, but even with double the efficiency of modern solar, I'd need 6m2 of panels for the daily (220km) commute. I think I'm an outlier, but I don't think the fit for many will be as promising as Aptera suggest (since lots of parking is underground, for example, you're not going to get charged while you're away from the car).
Can't fault them for trying though.
It's not quite that simple though. Some postcodes in Australia cover quite large geographical areas with lots of people, others cover large areas with a few people, and still others might cover just a few suburbs in Sydney or Melbourne.
Wikipedia - Postcodes in Australia gives a few examples of extremes.
Given those sorts of examples, most Australians probably won't think of a postcode as being able to directly identify people.
The trick here is to use Profiles in Edgium (I think it's called Profiles in Chrome too?) to separate out your personas. I have between 4 and 10 on the go at any one time - one for my company, and one each for every O365 or Azure customer.
Also, if you can get yourself onto the CSP program, you should be able to set up the tenants so you log in once, and have delegated access to the other tenancies; you can then switch between them using links at the top of the various admin portals.
Because if they support older devices, people who won't purchase new devices won't purchase any new devices and they might miss out on ten cents of advertising revenue. This is horrifically bad, because that ten cents of revenue might mean that the shareholders would need to spend 10c of their OWN MONEY for the 600 foot power yacht.
And apparently we can't have that.
So your contention here seems to be that a SME with a need for SQL Server to support a line of business application for 50 users on 4 cores in a VM or container should pay the same as a large multinational who want to support 5,000 users on bare metal and 128 cores.
I don't agree with that opinion at all.
If you don't want to pay the licenses, buy or build a solution that doesn't require them. Otherwise - it's a business, pay for the software.
I was given to understand that an EW is there on behalf of the court (and therefore "called" by the presiding judge), not either of the parties to the action. It seems that would be the only reasonable way to have someone who is impartial in their role as the EW (and specifically that's because they are _seen_ to be impartial as well).
Honestly, I've spent five weekends and countless nights trying to get K8s to run in a truly HA mode in my lab (3 masters, shared storage for volumes and 2+ workers). I've read dozens of K8s guides, 95% of which build an "HA" environment with a single cluster master accessible only from the master itself.
The other 5% are outdated (even though at least one was only 4 months old, it referenced a process that had been replaced two months prior to the guide date) or plain don't work as written.
At this point K8s feels like ivory tower academia desperately pretending it's open but rabidly protecting its IP so it *can't* be replicated on premise. That's not really a recipe for a future in which I want to be building IT.
That in turn means that I don't currently see containers as a way out of the cloud mess, in case it's not clear.
I rather suspect that admitting that is irrelevant - wouldn't it be the case that only the owner of the mobile device would have standing (in the legal sense) in a disagreement about the specific devices?
After all, Alice can't sue Bob for Bob breaking into Candace's computer, right?
Your information is out of date. Edge Stable has been available for several weeks (both v79 and v80 are in stable, and those map closely to Chromium versions with the same inflated numbers). While there are Dev and Canary channels (and I run in Dev) it's definitely wide release as of Jan 15.
And as far as I can see, Edgium doesn't send the X-Client-Data header (nor did it seem to have an equivalent for Microsoft properties so it seems to be MORE privacy conscious than Chrome).
I don't get the continued FUD about Microsoft hoovering up info - unlike Google where you don't pay for stuff (which supposedly means you're "the product", right) you do pay MS. And regardless of all the noise about it, I've never actually seen anyone show any data suggesting that MS really is copying everyone's hard disks to the cloud. Even when all the "privacy invasions" are left on.
's water music was referencing a song by Tom Lehrer, "Wernher von Braun".