The benchmark might seem like apples-to-oranges, but they both do templating & routing, which is what is being tested.
I also address this on the benchmark page:
Why compare a language to a framework?
1) THT takes a “batteries included” approach by providing essential web framework features.
2) Some might be worried that running a transpiled language would be too slow to run in production, and Laravel is arguably the best example of a production-ready PHP framework.
Also, yes it is essentially a new language. I don't expect anyone to rewrite large PHP applications in it, but people might consider it for new projects.
Strangely enough, many countries also have cultural laws around what languages can be used for subtitles and audio tracks. For example, IIRC there is a country in SA that doesn't allow non-Spanish subtitles if the primary audio is in Spanish.
There are all kinds of permutations like that which can look like arbitrarily bad UI decisions if you don't know the greater context.
They're still pretty poor UI decisions if you're implementing misfeatures for every country based on the limitations of some countries. The more harmful it is, and the more money you have, the less justifiable it is.
I'm fairly positive, at the scale of Netflix, and the relatively minor amount of UI work they need to do, they aren't struggling to implement country-specific features. More likely than not, the ball either got dropped, or the misfeature wasn't understood as one.
I can understand that it might not seem useful to you if you already have a stack you've invested in, or it just doesn't fit the kind of projects you are working on.
But I just wanted to respond to a few things. :)
> A) Come from a javascript background
It targets all web developers, who are all generally familiar with JavaScript. It aims to be another tool, not a total replacement for all projects.
> B) Want to deploy on cheap hosters where node isn't available?
Shared hosting is still a very popular way to run websites. And node is often overkill for basic CRUD sites.
> C) Don't care about performance
There is a big difference between not needing to support 100+ requests per second and "not caring". :) Out of the box, PHP is fast enough for a lot of apps, and the overhead of THT is not enough to be noticeable at that scale.
> don't mind the additional complexity of the transpiler
There are no build steps, so the extra complexity just involves running the initial setup script. After that, you drop in new files and update them like you would with PHP. Ideally, it won't feel more complex than any other framework.
> Can this even leverage the enormous PHP ecosystem?
This is on the roadmap. Though there are still plenty of projects that don't use 3rd party libraries and could be built with THT as it is today.
I personally have never needed to use composer, etc. in my projects, but it's still something I am looking to support.
Thanks for the response.. Wish you nothing but the best...
My comment about complexity in the transpiler is more about debugging and what do you do when you can't figure out where your bug is, because it's in the transpiler. ;)
And I think your last sentence regarding composer makes it clearer whom this is aimed at.
It's not just marketing fluff. It really does a good job of reflecting what it's like to work there, and is taken seriously by everyone from the top down.
As I mentioned elsewhere in this thread, it's not a competitive environment, and I've experience very little politics and drama. Those things are explicitly counter to the culture.
As for long hours, while the work day can be pretty intense, the office is usually empty by 5-6pm.
A while back, I remember reading something about how Netflix will periodically re-interview their employees and if they fail, fire them. Is it (still) a thing? Also, what about stories such as this: http://www.npr.org/2015/09/03/437291792/how-the-architect-of...
"Today, Netflix offers unlimited vacation time, high salaries. This company doesn't care how long you're in the office, but they do care about what you produce. And you have to produce. Here's another slide. We are a team, not a family. We are a pro sports team, not a kids recreational team.
Netflix fires people, even hard-working people.
HENN: How many people have you moved on?
MCCORD: Oh, hundreds.
HENN: I talked to a lot of former employees were OK with this deal. I mean, after all, Netflix pays well. But some talked about a culture of fear - always been worried about being fired. Patty says the company's culture is a big reason for its success. She says it keeps Netflix lean and nimble."
As a Netflix employee, I wouldn't really describe our culture as cutthroat at all. My experience in 5 years has almost entirely devoid of politics.
For example, from our culture page:
"We have no bell curves or rankings or quotas such as “cut the bottom 10% every year.” That would be detrimental to fostering collaboration, and is a simplistic, rules-based approach we would never support."
Our culture can feel intimidating, especially if you have any degree of imposter syndrome, but with very few exceptions, my colleagues and managers have always been supportive and really great to work with.
>>We have no bell curves or rankings or quotas such as “cut the bottom 10% every year.
This is possible in companies growing rapidly or bringing in lots of money.
In 1990s, It was very common in Indian IT services firms to openly talk about salaries and promotions, as getting the best of both was easy for every one. Enter 2000s and all of this came to an abrupt end, stack ranking emerged out of nothing as people competing for limited resources compared themselves with each other.
Its not that hard to pay and promote when things are going well. Its only when the scarcity kicks when the real issues start.
This is spot on! The same goes for so many other companies. When the pie is growing, everyone gets a bigger piece, there is lots of whitespace to move into, mid-level folks can stretch into more senior roles, faster and more effectively than it takes the company to hire external senior people, and so on.
Interesting. I have a few friends who have interviewed and talked with people there and it's always been described to me as a place that does cut the bottom performers every year. When you hiring only "rock stars" those bottom 10% may still be great.
I'm glad to hear that may not be true as it's always put me off from attempting to work there.
Do you trust a random commenter on HN on the Internet than a few of your own friends? This is something I wonder about. I weigh the opinion of random reddit posts than people I know in real life. How did I get that way?
Perhaps something here is that this is an identifiable person who actually works at netflix making a semi-permanent and public statement.
Contrasted to a friend who heard something passing on information that they were not expecting to be held accountable for.
I might trust my friends more, if my friend and someone on the internet had the same information, but might trust someone on the internet more if they had better knowledge.
Also, they did say "might not be true", and it certainly makes sense to use new knowledge to adjust the probabilities of belief.
> Do you trust a random commenter on HN on the Internet than a few of your own friends?
No and yes? The problem is I didn't have any friends who worked directly for Netflix this is mostly what they had heard from others and what one supposedly asked about during his interview a few years back.
So yeah I trust them but they also got their information second hand which, when compared to "stranger on the internet claiming to be working at Netflix" I don't see a huge difference in.
I imagine that it's like playing on a professional sports team. If you can't hack it in the NBA, NHL, NFL, etc, you'll know inside of you. Guys that make it to The Show have very little doubt in themselves as to what they're capable of doing, and even the superstars are mostly working their butts off to get even better, and that goes double for the benchwarmers. See any YouTube video on Kobe Bryant's work ethic. But even the benchwarmers know that they belong if they put in the work.
The benchmark might seem like apples-to-oranges, but they both do templating & routing, which is what is being tested.
I also address this on the benchmark page:
Why compare a language to a framework?
1) THT takes a “batteries included” approach by providing essential web framework features.
2) Some might be worried that running a transpiled language would be too slow to run in production, and Laravel is arguably the best example of a production-ready PHP framework.
Also, yes it is essentially a new language. I don't expect anyone to rewrite large PHP applications in it, but people might consider it for new projects.