Re: Web/Native: gap analysis

Hi Jo,

Le samedi 05 octobre 2013 à 12:27 +0100, Jo Rabin a écrit :
> I spent some quality time reading these documents this morning and
> think that fellow WebMobsters will find the time as rewarding as I
> did.

Thanks a lot for taking the time to read this and for providing
feedback! 

> While I let my thoughts brew further, some immediate reactions:
> 
> - Focus on the user is something very much to be welcomed. Identifying
> and distinguishing direct users of the technology - in the form of
> people who wish to build products and services - and indirect users -
> the targets of the first set is extremely important. I wonder if we
> should explicitly call out a third constituency, which is
> manufacturers and browser vendors. As the key enablers, identifying
> their concerns would be useful, in my view.

Yes, that might also be a fruitful perspective; I guess that given this
concerns a much smaller number of players (than users and providers), I
would tend to defer proper work on this to a later stage; but it feels
like at least sketching out what this would entail early on would likely
highlight some interesting questions, so I'll try to take a stab at
this.

> - It's probably worth noting that some of the concerns are in tension
> with each other. I'm particularly thinking of a reasonable desire to
> integrate more closely with the look and feel of a particular target
> delivery context (device) which would appear to be in tension with an
> equally sensible desire to provide a continuous, easy to navigate
> experience of the same product or service across platforms from
> different providers. 
> 
Definitely! I wonder if you have a suggestion as to where to highlight
that tension?

> - Needless to say, perhaps, I think that some aspects can be called
> out with more emphasis. Here I'm particularly thinking of best
> practice concerns relating to product development, especially reaching
> minimum viable product (MVP) and measuring use and iterating the
> product in a continuous feedback cycle. The Web has plenty of
> advantages here, and is particularly suitable for those who can go as
> far as doing A/B comparisons.
> 

This sounds indeed worth highlighting; would this mean adding "reaching
MVP" and "allowing quick iteration" in "development cost" in
http://w3c-webmob.github.io/gap-analysis/provider-comparison-framework.html ?

Could you describe the advantages that you see from experience for the
Web?


> - I wonder if we can emphasise areas in which the Web could have a
> distinct advantage. Resumption from the point the user last left off
> (resumption of task context) is noted as something that Web apps are
> possibly lacking. Transfer of the task context between delivery
> contexts is an area that the Web could explore as being something that
> is a unique and important value.
> 
I had tried to capture this under "Getting the content / configuration
of the app synchronized across devices" — do you think it's worth
calling out separately? Or clarifying?

> - Also, needless to say, I guess, one can argue the exact conclusions
> under some headings.

I think they can probably argued under many headings indeed :)

>  For example, I'm not clear that there are lots of effective Web
> development IDEs. To say that an IDE is effective, I think a clearer
> framework for evaluating the features offered against some kind of
> idea of what Web development current best practice consists of.

Yes, I think my assessment of quantity ("plenty of IDEs") as a strength
hides an assessment of their quality; I guess most of the feedback I've
personally heard about the difficulty of doing development for the Web
has focused on the difficulty to debug (in particular because the
browser remains a pretty big black box) more than on the difficulty to
write code; but I would be more than happy to elaborate on what weakness
exists in this space if you can share your experience there :)

>  It's my impression that there's a substantial body of Web developers
> who eschew IDEs in favour of top end text editors (Sublime Text, for
> example). From a designer rather than a developer perspective
> (ignoring for a moment any questions about the changing distinction
> between these roles) I'm not aware of a wide variety of tools in this
> area.

Clearly there should be more attention brought to design considerations
in my framework; it would be useful if you could suggest which "costs"
are worth highlighting.

And I must confess that I would not be in a very good position to assess
how the Web fares in their regard, so input on that would be great as
well!

Thanks again for your valuable input!

Dom

Received on Monday, 14 October 2013 10:11:40 UTC