Re: Notes on Web and Service Workers

> On 12 May 2015, at 17:17 , Brady Duga <duga@google.com> wrote:
> 
> I think, for purposes of replacing the packaging spec, the important point is the ability to handle requests for specific resources. So, if a web app loaded from http://foo.com/bar/myreader.html has a tag like <img src="img1.jpg">, the browser goes out and asks foo.com for /bar/img1.jpg. With service workers you can now have javascript intercept that request and fulfill it with local data, a completely different image, something from a package, etc. This replaces the packaging specs ability to have a single blob with all the data already local - instead of defining how a browser can load data from a specific cache, custom logic can be applied to every load which is a more general solution. For instance, compression and encryption can be handled by the service worker. It is also how web views on Android and iOS already work, which is why it is easier to handle this in native apps for those platforms. Basically it is exposing this useful behavior to the web as a whole.

Thanks for this concise addition, Brady!

> 
> Caveat: I am not an expert in this area, and have not read the specs.

Same here:-)

> I don't really know the implementation details, or how content origins, etc might impact all this.

And we should also remember that this is really bleeding edge stuff. Difficult to predict, but I believe we are still many months (years?) away before this is stable, though Chrome and Mozilla seems to come with the ability to switch these features on (they are off by default in the standard distributions). See also [1]

Ivan

[1] http://caniuse.com/#feat=serviceworkers

> 
> On Tue, May 12, 2015 at 7:16 AM Nick Ruffilo <nickruffilo@gmail.com> wrote:
> Ivan,
> 
> Web/Service workers can be some seriously complex topics, especially if you don't come from a computer science/linux background, but I'll try to explain them in completely "human" terms.
> 
> Imagine your web page - instead of being a bunch of HTML/CSS/Javascript, being a person - lets say ME.  (obviously this is a bit of a simplification, but you'll get the point) When you first access that website, I create outlines for the objects based on the HTML code.  Then, I paint and manipulate how those objects based on the way they look (that's the CSS).  I'm one person, so I can only do one thing at a time.
> 
> Next, once everything is laid down, I get to "work" (javascript).  As your mouse moves I check my list of instructions to see if I should be triggering an interaction, etc.  But I'm only 1 person.  Technically you can have me complete multiple tasks (have a clock that displays the time, play an animation, etc) but, for the most part, I have to figure out how to manage all of that with my arms and legs.
> 
> In comes "Web/Service workers."  Lets say I'm getting really tired, I can call a "Web Worker" (temporary labor) to come handle some tasks for me.  Since this is a new person, they can handle the tasks independent of what I am doing.  Since a web-worker is temporary labor, they will go away as soon as you close that page.
> 
> Then we get into Service Workers.  Service workers are like butlers - they stick around even if you leave the page.  So, if you ask them to make you a sandwich (because you're so hungry from running so much javascript) the user can close the page, then come back later, and the butler is ready with the sandwich.
> 
> Ultimately it comes down to THREADs.  A thread is a "line" of functions.  Most of programming is linear: Execute this code, then move on to the next.  Repeat.  The issue is that if one task takes a long time, everything AFTER that gets delayed.  Creating a 2nd thread means work can still get done.  Today's chips can easily handle multi-threaded operations and excel at it.  Web workers are a way to easily define threads, whereas in the past you had to do it in a very "hacky" way using timers (in javascript).
> 
> Hopefully this clears things up.  If anyone spots a flub by my explanation, please let me know!
> 
> On Tue, May 12, 2015 at 8:24 AM, Ivan Herman <ivan@w3.org> wrote:
> Tzviya, Markus,
> 
> I have spent some time trying to get *some* idea on what Web Workers and Service Workers are. They are spinning my head and I am sure there are many details that I simply did not understand. Nevertheless, I thought it is worthwhile jotting them down separately, because they might affect the whole EPUB-WEB discussion:
> 
> https://github.com/w3c/epubweb/wiki/Notes-on-Web-and-Service-Workers
> 
> I would be pleased if somebody on the group, who has a better understanding of these, could comment/update the file!
> 
> Ivan
> 
> ----
> Ivan Herman, W3C
> Digital Publishing Activity Lead
> Home: http://www.w3.org/People/Ivan/
> mobile: +31-641044153
> ORCID ID: http://orcid.org/0000-0003-0782-2704
> 
> 
> 
> 
> 
> 
> 
> --
> - Nick Ruffilo
> @NickRuffilo
> http://Aerbook.com
> http://ZenOfTechnology.com
> 


----
Ivan Herman, W3C
Digital Publishing Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: http://orcid.org/0000-0003-0782-2704