This is unbelievably cringey and this person is in the wrong profession. The fact that they seriously pretend that good programming = boring is embarrassing, as is their caricature of people who enjoy the field as cowboys.
>Screw writing anything down or discussing your issues with you further! This is exciting! Oh if it goes wrong I’ll just quit because look at my CV, I’m doing triple heart bypasses like crazy!
This person needs a hug or something. Many doctors absolutely love their job. They love doing things that matter, and dealing with unique problems (and almost all problems are unique). Their anecdote is yet again a tired jealous screed trying to imagine themselves to be the good coder.
They update quickly because they know every trivial update will be a front-page on Reddit as a bunch of Microsoft boosters try to shout "SEE WE'RE HERE TOO!"
This is just not right at all (and one is a document data format, the other an IPC, so the comparison is flawed to begin with), not to mention that you're focusing far too much on the XML aspect which was utterly irrelevant in this case -- MSXMLHTTP actually had nothing to do with XML.
But from a Microsoft shareholder's point of view it was a disaster.
As one of the very first outsiders to actually use MSXML in a project (with the Usenet archives to prove it), embracing it when the first betas came out...
...not really. The web was just starting to become active and interactive, and there was a huge pent up need for this, and actually dozens of disparate solutions. The most common was simply using a hidden frame (security controls were vastly simpler), or a Java applet for the comms. Microsoft provided a widely released solution just as the industry was bursting with need, so it go embraced, but if they didn't other options were quickly poised to fill the void.
It already pushes the boundaries of "obvious promo post presented as some sort of operational wisdom". I think they realized their limits and held off making it too obvious.
>Because moving an existing site to another platform is non-trivial.
It's not just that it's non-trivial, it's that when it comes down to a real analysis, these "clearly better" alternatives aren't clearly better at all. They generally much, much worse.
There is a hipsterism in technology just like every other field where people identify themselves by the things they reject. The GP post is such a claim, with over the top bombast making everyone know how much they dislike Wordpress, acting as if there is some unnatural force (e.g. user delusion) keeping it popular. It's popular because it may just be the best option for the purpose it serves.
And I don't develop Wordpress themes or plugins, and of course curse it occasionally. I leverage it as a tool.
I've run a sporadically popular (100s of thousands of uniques on some days) blog on Wordpress for years. All running on a single spot instance (if it was killed I have a script to roll over in minutes).
The database, Wordpress, the file storage. Entirely self contained.
I occasionally persist the AMI so I can spin up a new updated system base, and have regular backups of the files and database. My monthly cost is ~$9 on AWS. I've had literally minutes of downtime over the past several years.
Obviously this guide is geared to a more serious installation, but there is the danger that the more reliable a system becomes, often the less reliable it becomes. EFS and NLB configuration changes or problems taking the system offline, for instance.
HN has gone to shit, and one of the reasons is that huge numbers of .NET developers somehow funneled in and started directing the content, so absolutely terrible submissions like this one float to the top. As does every future-claim of Microsoft's, every bit of oversold shovelware, etc. It's rather incredible.
And yes, I realize this will be dead, and I am completely okay with that. I achieve my purpose posting in hidden silence.
When a company like Microsoft or Apple buys back shares, they retire them. This in effect gives every remaining shareholder a larger ownership stake. Which is why shareholders love buy backs if they don't think the company can do something better with the cash internally (e.g. R&D on a new teleporter).
In a perfect information and consideration world the market cap would reduce by the amount spent for the buyback: Company is worth $300B of pure market valuation, and has $100B of cash and equities, so the capitalization is $400B in this hypothetical world, and if they bought back $100B of shares and retired them, it would conceivably drop to $300B.
The market doesn't do that, however. And if it did, the person you replied to would be arguing exactly the wrong thing, as in 1999 Microsoft had a cash pile of just $15B. Today they have a cash pile of $100B. So the differential in market caps would be even more substantial, not less.
In reality the capitalization is based almost entirely on anticipated earnings and prospects, and whether the company has $1B or $100B is immaterial, leading to the surreal situation where companies have more book value than market capitalization.
Your assumption is completely and absolutely wrong. Not just wrong, but actually the opposite of the traditional reality -- market cap usually doesn't even consider cash and short term equity reserves for traditional companies, so you end up with companies that have more book value than market value. Which is exactly why these companies buy back shares -- a smaller circulation benefits their shareholders more than money sitting in an account on the company books.
These tests churn GBs of data (necessary given how fast flash is), rendering the cache largely irrelevant. As I mentioned elsewhere, it is incredibly unlikely that parallelism plays a part (and has never been the case in prior devices, nor is it the case with the vast majority of SSDs. Anyone blindly buying larger for more parallelism is very likely to be disappointed). Different vendors make their memory solutions, and there are huge variations in options. We've seen this across a single model of a set size when two different vendors provide the storage.
In the end it really won't make a difference for real world use.
That isn't a general rule at all, and it would be foolhardy to generally apply it. While there are cases where it may be true, in many other cases multiple chips sit behind a single controller and are accessed serially -- some easy examples are that the Intel 750 1.2GB has virtually identical performance to the 400GB. The Samsung Pro 950 256GB, 512GB, and 2TB all have virtually identical performance.
And on smartphones, it is very, very seldom the case that flash storage varies in speed between capacities, beyond simply vendor differences. Which is probably the case here.