It's definitely hard. There was certainly no clear cut solution to any of the problems I included on the card in the picture. We evaluated 8 or 10 different solutions for out of the box stuff, found things we liked and didn't like about all of them and eventually decided it was best to build our own. At each step there was a lot of debate because we knew this would probably be used beyond just the Careers project and be used in the Q&A project as well (separate at the time), so it wasn't just worrying about my team's concerns, but everyone's. We won some debates, we lost some debates, it was very hard.
The localization project I use as an example here was definitely one of my top 3 hardest projects of all time (and that was years ago). It was not a particularly difficult technical challenge, it was difficult because it touched every single aspect of the codebase. The project took me and a coworker 3 months to build out the infrastructure for, then another 3 months of actually rewriting everything to use it, and explaining to every single other developer about why we made the decisions we made in order to teach them the different ways they were going to have to write code from now on. Social challenges of the workplace are hard; we're not always looking for technical difficulty.
Thanks for posting this. One thing I didn't write about is the objectiveness of interviews (felt like a different topic). Just about any question can be okay as long as it follows a few simple rules. 1) if you ask a question, you should have an expectation of what you're looking for as a result in some quantifiable way and 2) how does that result affect your decision to hire or not. The worst interviewers usually don't have an answer for #1 there and they probably decided to hire you or not very soon after you opened your mouth for the first time. This article fails to state why each of these questions is valuable and what you should be looking for as a result of asking any of them. With the addition of that, it would be a much better article.
I'm familiar with the LLVM direct to wasm tool, but I was unable to get it working using the steps from the emscripten docs. The binaryen toolchain is pretty easy to digest, but I've been using the emsdk almost exclusively for the convenience. Once you have to rework part of it from source, I just don't know enough about it to debug it yet.
Thanks for the info on where to find outstanding proposals.
Initially my thought was to compare it to the first compiler, back in the days when punch cards were the interface, but I decided against any history lessons that go that far back. In that analogy, the byte code is the machine, and now we all need to be writing compilers for it. This is why so many languages compile to javascript, it's the assembly of the web (not my trademark), in the browser world at least. If you want this code to run in that browser, you gotta pay the penalty of cross compilation inefficiency.
I actually think being browser agnostic is more important; it is sort of the reason we're even looking at wasm replacing asm.js in the first place. asm.js is still javascript, even though it gets handled differently under the covers. Having a second option of language supported by all major browsers is a huge deal, especially one with static typing.
As for the naming thing, a little bit of bikeshedding is healthy from time to time.
We (at Stack Overflow) have a new documentation product coming down the pipe with an emphasis on examples and I thought it would be awesome if we could expand our stack snippets feature to languages beyond javascript. A reproducible example right there on the webpage. In researching existing solutions, nothing was satisfying, and then I had the idea to just use asm.js and write an MSIL interpreter. I spoke with some people familiar with asm.js and decided that it wasn't particularly difficult, but it had a ton of limitations from js file sizes to ways to get all of the libraries people ask about uploaded, and of course regular breaking changes in the browser. Wasm seemed like it could help with some of the issues being a formal web specification with representatives from multiple browsers.
Other than that, I can see a lot of uses for it, from small neural nets in the browser to high performance applications like photo editing, and of course gaming. It might not be for all applications, but it certainly has a spot.
That's fair. Originally asm.js was named capsule.js but that was taken and didn't really mean what it was. We could also call it modules but there is another very related spec out there for that as well. What would you call it instead?
I had the exact same problem when I first started. Someone once told me "what we can throw away to solve this problem", so I kept removing parts until I got here which I think is currently the easiest way to get started.
The last 10 minutes where you ask questions is a bad time to waste asking someone to judge your 3 year plan when they just met you. Most of these questions are better asked when you fail an interview and request feedback as to why.
I specifically state when I perform interviews that I will not answer the question "How well would you say I did in this interview?" The interview isn't over yet, and that's impossible to answer.
You should really spend this time to see how much you would want to work at the company. Ask about their processes, the things that would absolutely turn you off to a company or the things that would absolutely turn you on to a company and make you forget about all their other bullshit. Ask about the problems they are trying to solve. Ask about how smart the other people are who you'll be working with.
Don't waste time in the interview checking if you passed. You'll know very soon anyway and there is always the chance to get this feedback afterward.
> What are ‘presentation semantics’? I think this is a corruption of the concept of semantics. Semantics are what a thing is, or means, or does; not how it is presented
Semantics just means to give something meaning. Presentation semantics (http://en.wikipedia.org/wiki/Presentation_semantics) are just marking up the data to declare your intent for how you would like it to be presented.
> As I said, I don't believe this is an appropriate approach for a medium which has no absolute, definitive interpretation. A website isn't how it looks, because how it looks will differ depending on how you access it. The only absolute truth when talking about websites is the semantic purpose of the markup underlying it, because purpose is universal: a navigation element is a navigation element regardless of whether it's being displayed on a TV, a desktop browser, a smartphone, or being read out by a screen reader.
A HTML document is just text. Technically adding links, particularly navigation, is just presentation markup. As you said, it's up to the interpreter to make sense of that markup. The basis of your argument is that presentational markup is bad though, so I'm left unconvinced here.
> Changing how you structure the CSS in your application based on the whims of people whose area of expertise isn't CSS sounds like a nightmare scenario to me. If you're bootstrapping with a small team I can understand wanting to have everyone being able to contribute, but you should also be able to recognise what the limitations of their skills are. In the example you give, I would be using my web developer tools / inspector etc. to find out the origin of the erroneous property. If you're using a preprocessor there are source maps that can help with identifying how a rule was built. I certainly wouldn't conclude it's easier to use a presentational class name just so that my CTO could understand the code.
I'm concluding that presentational class names are easier to fix for the regular small changes that come down the pipe in our project. Even if you're regularly making big presentational changes to your entire site, then you'd need something a little more abstract, like using a class to declare an element as a button instead of the direct color it should be, but like I'm said, the argument against presentation markup is still unconvincing in the face of simple maintenance.
Let me start by saying I am the most vocal person about semantic HTML at my company, but it really depends on your purpose, and I'm not sure you are realizing you are taking a hard line. Also, I would have preferred you used the 90% of your words to give better examples of how to do things better than overstating how OOCSS and BEM are wrong. Well developed solutions are always better than just explaining what's wrong unless there is no solution and the existing proposals are wholly unsatisfactory.
Take a step back and realize that there is at least one more way of doing this. Why not have the presentation be the specification, and we can just build the html to conform to the that specification instead (via the use of helpers and all other commonly accepted forms of duplication reduction). Your post basically declares this Doing It Wrong™, but I imagine that's just because you'd rather see better looking html than you would see better looking css.
You have the semantic issues well thought out, however you are flat out declaring data semantics are more important than presentation semantics. For most projects, the HTML is only used in a browser, and isn't the source of truth for the application, i.e. it is inherently coupled with the presentation. They might as well be one and the same because this is it's only purpose in the world. Leading from that, the only issue is maintainability, and here is where your ideas aren't fully developed.
As will all separation of concerns, your solution (the one about breaking news) also adds a layer of misdirection. Sure I don't have to change the html to change the presentation, but now I have figure out where that part of the presentation is modified. My CTO and designer (2 people I have a lot of respect for) were the people who eventually changed my mind on this. They said, "I don't know why this button is blue, but it's supposed to be green". Well it turns out it's blue because it has the class blue on it. Instead of trying to figure out what selector was causing this button to be blue, all I had to do was remove the blue class and add the green class. It took 15 seconds. It took probably 3 or 4 weeks for me to relent on this before I accepted, but it is much easier for everyone (think of the new hires as you pointed out!) if they don't have to know tons of info about the domain, but just look for the offending element and make a simple change in place rather than have to unravel the css toolkit to find where my button was made blue instead of green. Additionally, they know this change is only affecting this one element they were told to change. On our team, that is better maintainability because that is how real changes are specified. In your case, by the time this spec makes it's way to the developer, it needs to have traveled through people who know that part of the web page is a breaking news container and translated from whatever the original complaint was or the developer will likely have another problem to figure out before he can even begin changing what will likely end up being a single css rule.
Loose coupling is simply about defining avenues of change, and it only improves maintainability when it enables making changes easier. This has been extrapolated to mean "does this have to change or not", which is over thinking it a bit. In my example, I made a typical update by changing exactly one thing in exactly one place, without having to traverse any levels of misdirection. Just figuring out how to make one thing not have to change doesn't make it more maintainable.
When we're talking about web pages, it's also important to realize that we're typically not changing just the presentation of static data; we're adding and removing features to figure out what makes this page more useful. The HTML is typically going to change anyway when we're modifying the page, which again leans in favor of a common set of CSS selectors to build against (in terms of avenues of change).
As I said, I'm probably the most outspoken person at my office when it comes to semantic HTML, but it's only when I'm describing actual data, i.e. the parts you read. These parts tend to change very little and can be generated and stored in their already rendered state. The parts that do change more often gain a lot from the benefits of OOCSS and BEM techniques.
Sure there are a few questions with highly voted up answers because they are widely useful, but there is certainly the minority of high rep earners. Even the new people who are gaining lots of rep are doing it by answering hundreds of questions.
Yes, with a strong bias to not forking and maintaining. Stale project, well you gotta fork and maintain. Active project, avoid forking with the intention of using a custom build in your system.
Were they actively making changes to the project and running custom builds? If not then they just had nothing to contribute to the project and there is certainly nothing to feel disgusted about.
This is exactly the pain my advice helps you avoid. The implicit cost of creating a fork is that future changes higher up the chain could either cause you to be stuck at this revision or give you an additional bulk of work in order to update. Of course anyone can update and run a custom build for their project, but the cost is often not worth it.
It should be noted that the situation where you intend to fork and maintain the project, this does not apply, and by all means make your changes. In the case where you intend to update, this pattern will help you avoid the pain.