Would that mean you could never take Uber seriously as an technology company again? I suppose under certain circumstances a company does deserve a death sentence of sorts in the court of public opinion.
Your attitude includes an implicit "trust the smart people."
But the smart people are proving again and again that they are not worthy of our trust. They find bigger and better and stealthier ways to learn about us so they can influence and move against us in the future. Why should we trust anyone who sees us as a product?
You lost me on how you went from "They find bigger and better and stealthier ways to learn about us" (which I agree with) to "move against us in the future". Care to elaborate?
Or to put it a bit more shortly, extreme data collection is a trojan horse, for a brave new world where individual taste, free will, and unpredictability are nearly completely quantified.
Lots of people home school their kids, grow their own produce, and kill their own cows. One doesn't need to do everything themselves to want to do some things themselves. And if your premise is that one needs to be an expert, before doing anything on their own, then how would people learn? How do people become experts?
The implication in your original comment is quite clear, "why don't you let experts do what they do and you spend your time doing what you do?". So, you actually did say that...
But to suggest that everyone should do (or will do it, as some people are commenting here) it is bananas.
The comment you replied to didn't suggest that everyone should self host, it simply stated a preference. Your comment, however, was the exact opposite, and suggested people shouldn't self host "unless they do servers"
Sometimes, yes. And in each of those cases, the right and motivated person can produce a better product that institutions.
Personally, I self host services on IaaS (Vultr). At least then I have full ownership of my data, and it guarantees I'm running on open source software that can be modified for data extraction at any time.
I do the same. And some times it goes down and I'm busy so it stays down for a day. And I never find time to keep everything updated to the latest security patches, etc. Fun hobby, but I wouldn't trust that as my only repository of data.
You do own your own phone right? Owning a server on the cloud is no different. There will be a future shortly where everyone keeps their data on their own servers instead of handing out to all these corporations.
You should read the terms of service with your Cloud provider again. When you go out and buy a phone with cash, nobody can take that from you. When you go out and spend money on Cloud servers you will not walk away with that data should you stop paying your bill, or the company is hacked, or there is a scandal and they go out of business, or they decide you're breaking their rules and they delete it. When you setup an externally available NAS from home the only thing stopping you from accessing it is your ISP and your electric company, and those are already common denominators of accessing any Cloud storage medium. And lets not forget Uncle Sam and random Google execs picking up the hood every so often to see what's under there. It's a lot of work, and it isn't for everyone, but the benefits outweigh the risks for me.
OK, since you own your phone. Do you have any clue about security and availability of your phone? Why do you think millions can own and "run" their own phone but not own and run their server?
Sure. And that's called a hobby. You can do whatever you want, but the point here is this is not what the masses are going to do. Or should do, for that matter.
To be fair, the title wasn't "New study proves that..." or even "Scientists say...". It was very clear that this a blog post written by a dude sharing his own experiences and anecdotes. Not some "proof" that life is better with less hours.
I'm confused about what the goal of this is. The best thing about PHP is its dynamic nature, and the fact that any new page load is a completely new universe. Why would you want to loose that and compile it to a, well, compiled language?
The subheading for their blog post summarizes it nicely:
"In our last blog post we discussed some of the advantages and shortcomings of PHP and the .NET framework, and pointed out why the two should not be compared with each other. Today, we are introducing a tool that can bridge the gap between these two frameworks, enable the developer to produce code in PHP that is both-way interoperable with .NET, and therefore make use of the specific advantages of both, PHP and the .NET framework:"
There are already compiled versions of PHP, like HHVM. Compiling a program does not preclude it from being dynamic. Also, I'd argue that throwing out everything with each page load is throwing out the baby with the bathwater. It's possible to create new workers for each task which don't share any sort of state. So you can both precompile and destroy the world with each instance.
So you can both precompile and destroy the world with each instance.
I'm pretty sure that's what PHP 5 does already with the opcache.
Looking at non-jit engines, I'm more familiar with jscript/vbscript under active server pages than I am with PHP. But even that engine from 1998 caches the "compiled" bytecode of the page in a scripting engine that can be run on any thread, and can be cloned. Once the engine is finished running the current request, it drops the state from the request and awaits a new request with new state. It never destroys the scripting engine unless an actual file changes. I'm pretty sure that PHP 5.5 is similar in that regard (though I'm certainly not 100%).
So when talking about 'throwing out everything' with each page load, I'm pretty sure that most people are talking about the actual state from the request, not bytecode cache. Compare this with something like nodejs where it's possible to reference a global variable from inside a request handler, and thus have a possibly hidden global dependency. Whereas with ASP/PHP, it's just not possible without making concerted efforts to call out to non-script land.
Specifically, for a non-jitted language, a very reasonable concession to performance is to go ahead and force each request to run in it's own world, and then 'destroy the world', because at that point you can do huge amounts of clean up since you can make a guarantee that no other request could possibly use that request's state.
VERY opinionated! You are going to get downvoted to heck for comments like that.
Try personalizing it: 'I found PHP much easier to learn for ..... but when I wanted xyz performance, .NET performance was abc.'
Give a use case / example where no code changes helped.
As others have said, apparently PHP7 is better than PHP5...
You are (VERY!) unlikely to change the minds of die hard .NET or PHP people, but you'll have a lot more potential with someone who doesn't know either...
I totally empathize with people that are truly affected by words or ideas. What I can't tolerate is people trying to impose me what I should be offended by.
I realize it's an inflammatory article (Harvard and Yale having their differences, what else is new) but it's definitely a unique take on this phenomenon from a historical perspective.
Dumb, maybe, or maybe not, but she's definitely a lying sociopath, no history of successful leadership, doesn't answer questions, and yet thinks she can be President. I saw her on Meet the Press last week and I was genuinely afraid, creeped out even.
> Not everyone currently running for president is doing so because they believe they actually have a chance of becoming president.
A lot of people seem to miss this key detail.
These people have advisors, analysts, and so on. They know that their chance of winning is extremely slim. However they don't care because the entire point may be:
- VP slot.
- Increase public image in general (for future runs for house seats, book deals, TV appearances, speaking opportunities, etc).
- Pushing a specific political agenda into the debates.
The unnecessary bashing of PHP, the lame justification for a sloppy architecture, the immense ego... I don't think hack culture should be used to justify asshole-ery.
