Mailing-List: contact flex-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: flex-dev@incubator.apache.org
Received-SPF: pass (nike.apache.org: domain of devudesign@gmail.com designates
 209.85.215.170 as permitted sender)
Message-ID: <50D38A85.8040102@gmail.com>
Date: Thu, 20 Dec 2012 22:00:37 +0000
From: Daniel Wasilewski <devudesign@gmail.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64;
 rv:17.0) Gecko/17.0 Thunderbird/17.0
MIME-Version: 1.0
To: flex-dev@incubator.apache.org
Subject: Re: [FalconJx] New JavaScript runtime format prototype
References: 
 <CAD8jDGL2Fu_sdkHq1Juism5P1jCTeY92SA2Np2fCN=-TcgTG1Q@mail.gmail.com>
 <20121220132355.15277ggwgnbmfe7v@franklin.liquidweb.com>
 <CAJs+wW3Oc_ty0Ew73sjzXiMN1Jdgkz=WMTgTiChUet4BA6sw_Q@mail.gmail.com>
 <CAD8jDGJ7iTdZvJivMC4H85ykuQGKod9SSoAzZocCWLvg8g=RRQ@mail.gmail.com>
 <CAJs+wW2j+wq4J-0Tx8ZLB5r-aY0iJ5ry6y6kSWc8S73Zt63tPw@mail.gmail.com>
In-Reply-To: 
 <CAJs+wW2j+wq4J-0Tx8ZLB5r-aY0iJ5ry6y6kSWc8S73Zt63tPw@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

On 12/20/2012 9:54 PM, Erik de Bruin wrote:
> *tool/vendor dependency:* The Google Closure Tools are Open Source,
> available under the Apache License 2.0.
>
> *separation of concerns:*
> 1) a matter of choice and convenience, I'm not married to "goog" here,
> but I haven't seen a good reason NOT to use it, so far.
> 2) you are probably correct from a purely language translation point
> of view. But the "ultimate" goal here is not only a language
> translation, but also a framework translation. And I'm sure "some
> polyfills" are not enough to create a cross browser UI framework.
> 3) I'm hearing "different", I'm not sure I hear "better". Why wouldn't
> we use the Closure Builder for this? I explain the virtue of this tool
> in an earlier email and you can see it in action in the Publisher tool
> I set up in the 'develop' branch of the "asjs" section of the SVN
> repo.
> 4) the Closure Builder does a bit more than linking. Please see my
> previous remark and email, as well as the documentation available
> online.
> 5) GCC is also more than just an minifier. And we can run it straight
> from the cross compiler, as Mike suggested. That leaves us with the
> option of keeping the intermediate JS code as an supplement output.
>
> *testing:* never implied that GCC or anything would free us from
> testing any part of the code. And creating optimised/minified code
> straight from the cross compiler - without using an external tool as
> you seem to imply - would mean that we'd have to write our own
> optimiser/minifier into it. I'm not familiar with creating such a
> tool, but that sounds like a lot of work.
>
> *source maps:* I don't know what you mean by this, so I can't argue
> either direction :-)
>
> Like I emphasised before, and I'll repeat it every time it comes up:
> there is, at least not at this point in the development, an absolutely
> right nor absolutely wrong approach. I think we should look for the
> best of both approaches. Our approaches are essentially similar (only
> the tools differ) and I think both have strengths and weaknesses.
> Let's work on this a bit and than revisit this discussion and decide
> on the merits which one fullfils which requirements best.
>
> Have a great holiday, and keep having fun!
>
> EdB
>

I am signing off under Erik post and his thoughts