W3C home > Mailing lists > Public > public-xformsusers@w3.org > October 2019

Re: On the Expressive Power of Declarative Constructs in Interactive Document Scripts

From: Steven Pemberton <steven.pemberton@cwi.nl>
Date: Tue, 15 Oct 2019 10:59:22 +0200
To: "Guntur Wiseno Putra" <gsenopu@gmail.com>
Cc: XForms <public-xformsusers@w3.org>
Message-ID: <op.z9o1w8bvsmjzpq@steven-xps13>
The word 'model' in XForms refers back to the model-view-controller (MVC)  
paradigm that originally appeared in Smalltalk. However, in XForms the  
idea is somewhat more generalised: in MVC the relationship between model  
and view is one-way (from the model to the view) and the controller is  
responsible for the flow in the other direction. In XForms the  
relationship is two-way, with constraints and invariants achieving much of  
what the controller would have been needed for, although Events and  
Actions allow you to add your own effects where they are not supplied  
automatically by the system.

In retrospect, the word Form might have been a good choice instead of  
Model, in the sense of Form and Content.

Steven


On Mon, 14 Oct 2019 19:19:53 +0200, Guntur Wiseno Putra  
<gsenopu@gmail.com> wrote:

> Dear XFormsUsers and Steven,
>
>
>
> It may be inspiring:
>
> So it is about "model"...? as "the word is used in so many different  
> ways in common parlance as well as in >academia" (Patterson, Z.,   
> "Model", 2008: discussing the word in relation with social science) ...?
>
> https://www.espacestemps.net/articles/model/
>
> Until the MarkupUK 2019 it is still said that the components of XForms  
> are the model and the human interface (Steven >Pemberton, "Declarative  
> Applications").
>
> https://homepages.cwi.nl/~steven/Talks/2019/06-07-markup/
>
> Regard,
> Guntur Wiseno Putra
>
> Pada Rabu, 09 Oktober 2019, Guntur Wiseno Putra <gsenopu@gmail.com>  
> menulis:
>> Dear XForms Users & Steven,
>>
>> To share what may be inspiring (May we say what are below...?):
>>
>> Somewhere a city of networks, those networks of languages, ones learn  
>> on how to navigate it, how to work it out by such a >>strategic spatial  
>> planning: thus there is a multiplanar methodology...
>>
>> https://www.espacestemps.net/en/articles/strategic-navigation/
>>
>>
>> Regard,
>> Guntur Wiseno Putra
>>
>> Pada Rabu, 02 Oktober 2019, Guntur Wiseno Putra <gsenopu@gmail.com>  
>> menulis:
>>> Dear XFormsUsers and Steven,
>>>
>>> XForms, Networks of Languages, and Architecture...
>>>
>>>
>>> As we are trying to say architecturally about "XForms" regarding  
>>> with"networks of languages":  may we imagine such >>>buildings "Plan  
>>> of Pope  Sixtus V for Rome in Italy,1585", "Yi Yuan (Garden of  
>>> Contentment) in Suzhou, China, 19th >>>century" and "Plan for  
>>> Washington D.C., USA, 1792" with their network configurations of the  
>>> path (Ching, F.D.K, >>>"Architecture: Form, Space and Order", John  
>>> Wiley & Sons, Inc., 2007, pp. 276-277)...?
>>>
>>>
>>>
>>> Regard,
>>> Guntur Wiseno Putra
>>>
>>> Pada Rabu, 02 Oktober 2019, Steven Pemberton <steven.pemberton@cwi.nl>  
>>> menulis:
>>>> On Tue, 01 Oct 2019 17:32:50 +0200, Guntur Wiseno Putra  
>>>> <gsenopu@gmail.com> wrote:
>>>>
>>>>> Dear XForm Users and Steven,
>>>>>To share what may be inspiring:
>>>>>
>>>>> It is known that there are architectures of machines and systems  
>>>>> regarding with computing technologies: does it >>>>>sound fantastic  
>>>>> if there is a language supporting those architectures...? --a  
>>>>> language by which we may work out the >>>>>architectures...? --thus  
>>>>> we may build or renovate machines and systems using the language...?
>>>>>
>>>>> Of a reading, "architecture" consists elements "form", "space", and  
>>>>> "order": does XForm language -- together with, >>>>>if there are,   
>>>>> XSpace and XOrder-- embody part of such an architectural  
>>>>> programme...? --or at least potentially...?
>>>>
>>>> In XForms, the form is provided by the model, the order by the  
>>>> content in the body, and the space by the CSS.
>>>>
>>>> Best wishes,
>>>>
>>>> Steven
>>>>
>>>>>
>>>>>
>>>>> Regard,
>>>>> Guntur Wiseno Putra
>>>>>
>>>>> Pada Selasa, 01 Oktober 2019, Steven Pemberton  
>>>>> <steven.pemberton@cwi.nl> menulis:
>>>>>> It struck me that we should be making a collection of references to  
>>>>>> all papers about XForms.
>>>>>>
>>>>>> Please reply to this message with examples you know that should be  
>>>>>> included. I will collect them all together.
>>>>>>
>>>>>> Thanks!
>>>>>>
>>>>>> Steven
>>>>>>
>>>>>>
>>>>>> On Tue, 01 Oct 2019 15:40:30 +0200, Steven Pemberton  
>>>>>> <steven.pemberton@cwi.nl> wrote:
>>>>>>
>>>>>>> By John Boyer.
>>>>>>>
>>>>>>> Contains an XForms implementation of quicksort.
>>>>>>>
>>>>>>> ABSTRACT
>>>>>>> It is difficult to generally compare the succinctness of  
>>>>>>> declarative
>>>>>>> versus imperative programming as source code size varies. In  
>>>>>>> imperative programs, basic operations have constant >>>>>>>cost,  
>>>>>>> but they
>>>>>>> tend to be more verbose than declarative programs, which increases
>>>>>>> the potential for defects. This paper presents a novel approach  
>>>>>>> for a
>>>>>>> generalized comparison by transforming the problem into comparing  
>>>>>>> executed code size of a benchmark imperative >>>>>>>algorithm with
>>>>>>> a partially declarative variant of the same algorithm. This allows
>>>>>>> input size variation to substitute for source code size variation.  
>>>>>>> For
>>>>>>> implementation, we use a multiparadigm language called XForms
>>>>>>> that contains both declarative XPath expressions and imperative
>>>>>>> script actions for interacting with XML data within web and office
>>>>>>> documents. A novel partially declarative variant of the quicksort  
>>>>>>> is
>>>>>>> presented. Amortized analysis shows that onlyO(n) imperative  
>>>>>>> actions are executed, so the expressive power of >>>>>>>the  
>>>>>>> declarative constructs is at least Ω(logn). In general,  
>>>>>>> declarative constructs can
>>>>>>> have an order of magnitude expressive power advantage compared
>>>>>>> with only using basic imperative operations. The performance cost
>>>>>>> factor of the expressive power advantage was determined to be
>>>>>>> O(log2 n) based on a novel dynamic projection from the generalized  
>>>>>>> tree structure of XML data to a height >>>>>>>balanced binary tree.
>>>>>>>
>>>>>>> https://dl.acm.org/results.cfm?within=owners.owner>>>>>>>%3DHOSTED&srt=_score&query=10.1145%2F3342558.3345397&Go.x=0&Go.y=0
Received on Tuesday, 15 October 2019 08:59:30 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:37:52 UTC