Re: Last Call comments

On Mon, Apr 8, 2013 at 11:45 AM, Konstantinov Sergey
<twirl@yandex-team.ru>wrote:

> 08.04.2013, 19:24, "Scott González" <scott.gonzalez@gmail.com>:
> > On Mon, Apr 8, 2013 at 11:06 AM, Konstantinov Sergey <
> twirl@yandex-team.ru> wrote:
> >> Okay, look at the hammer.js pointer code:
> https://github.com/EightMedia/hammer.js/blob/master/src/pointerevent.js
> >> 55 lines of code are just redundant and is used to handle the simpliest
> things: the number of touches and the positions of touches. And this code
> will be completely broken if some other script on the page uses the
> capturing.
> >
> > I think it's unfair to point to an abstraction over several event types
> when that's exactly what Pointer Events is. If Hammer.js was coding over
> the Pointer Events API and mapping Touch Events to Pointer Events, the code
> would be very different.
>
> This code doesn't map anything. That's just the common counting pointers
> problem which will be implemented by everyone.
>

This one file alone doesn't really show anything. This is heavily tied into
an abstraction built by Hammer.js. If this were purely pointer counting
code in entirety, then the code would be self-contained. It's not. It
contains methods that are used by some other file in order to do the
counting. There is clearly more going on than just counting pointers in
that code. For example, the methods matchType() and getEvents() are clearly
unnecessary in a Pointer Events only environment. The updatePointer()
method wouldn't need a line to abstract around the identifier property and
wouldn't need a line to return a value. The getTouchList() method can just
use array.map() and cut the implementation in half. So now 55 lines is
probably around 10 lines. But even that number doesn't have much meaning
since this is clearly tied into other code to do the management around the
events themselves.

>
> >> We are forcing EVERY web-developer to write such code; we (Yandex Maps
> API team) wrote such code too - just because such simple things are "out of
> scope" and "we will provide helpers in v2". Meanwhile every single js
> framework will be forced to write that code.
> >
> > Surely that's not true. Most web developers will never write such code.
> They will simply use someone else's code which already does this.
>
> In first, there are lots of code which doesn't rely on frameworks.
> Cartographical APIs, the different "smooth scroll" plugins, etc.
>

Sure, but let's avoid obvious hyperboles. You said EVERY, I didn't say none.

In second, any errors in this code will be repeated hundreds of times.
>

Are you saying that it will by copy-pastd hundreds of times and the
original source won't be used? If the actual source is used, then the fix
is also deployed to all sites easily.


> In third, "not so many developers need to write the same code" doesn't
> answer the question why we are forcing to write such code at all.
>

No, it doesn't. But you continue to ignore the answer which is that many
developers want it.

> Also, what's wrong with iteration? Why is it imperative that v1 contain
> everything?
>
> Then between first and second iteration hundreds of developers will write
> thousands lines of code which will be thrown away at the V2 spec release.
>

Please stop with the hyperbole. There won't be hundreds of developers
writing thousands of lines of code. All of jQuery core is only thousands of
lines of code. All of hammer.js is 1,217 SLOC. So realistically, we're
talking about a handful of developers writing a few hundred lines of code.

Maybe we should answer then why to release v1 spec if it obviously lacks
> the mechanisms which are needed by every developer?
>

Because it's clearly not needed by every developer. Not even close.