Re: Developer tools

A small, late and somewhat random contribution on this topic.

Standards without conformance test suites tend to be implemented in inconsistent ways.

Things that are deliberately constructed for ease of learning and prevention of unintended consequences make for more reliable outcomes.

Debugging is a lot easier now that processors have specific support for debugging.

Presumably, in order to be "competitive" the Web needs to have the same kind of production lifecycle support as other environments - including debugging, memory usage, performance, unit testing, code coverage and so on …

The standardisation process within W3C - which usually seems to be solely feature driven - does seem to be changing to espouse conformance as a first class objective. Maybe thinking about other things such as having the right hooks for tools, accessibility (in the sense of being easy to learn) and other things - although not mobile specific - might be worth thinking about? 

Jo



On 19 Apr 2013, at 11:02, Dominique Hazael-Massieux <dom@w3.org> wrote:

> Hi,
> 
> Another remark that I've heard mentioned by big-projects trying to
> develop on top of HTML5 was the lack of good developer tools.
> 
> As an example, LinkedIn recently announced they were moving a way from
> HTML5 as their basis for their mobile apps, and the main reason they
> cited was the lack of debugging tools (e.g. to track memory usage):
> http://venturebeat.com/2013/04/17/linkedin-mobile-web-breakup/
> 
> Are there steps that W3C could take to help the ecosystem of developer
> tools? Are there particular barriers in our technologies that need to be
> lifted to make it possible for these tools to appear? 
> 
> Regarding memory usage specifically, the topic seems to have been
> brought up during the Web Performance Workshop back in November, but
> rather inconclusively:
>        Most memory information is not really useful to developers, as
>        it may include total system memory (and developers don't know
>        what other applications are running at that time) and different
>        machines have different characteristics. We may want to better
>        understand what the problem we are trying to solve and then work
>        towards solutions. 
> http://www.w3.org/2012/11/performance-workshop/report.html#i10
> 
>> From what I understand, this was looking at addressing the problem
> mostly in deployment stage, not in development stage, so maybe there is
> a lower hanging fruit in this particular space?
> 
> Anyway, feedback on what if anything we could do here would be very
> welcomed :)
> 
> Dom
> 
> 
>