Claims put to the test.

[richmond62]

Re Xavvi it is claimed:

For a simple app you won’t need to edit any code.

https://livecode.com/xavvi/

Now, I, by way of earning my bread and cheese run an ESL (English as a Second Language) school in Bulgaria.

I have a fair few computers running Xubuntu Linux on which I deploy LiveCode standalones of a fairly basic nature for children (6 - 18) to 'exercise' their English on.

Over an 18 year period I have built up a 'library' of about 60 of these standalones made on various versions of RR/LC from 2.2.1 to 9.6.3.

In very few cases do they not require the most simple programming imaginable: they DO take on average 8-12 hours to make, not because of the programming, but because of the media (images) involved.

I am going to lead readers of this thread through my latest standalone for my school:

[All screenshots have been reduced to 60 per cent of their original size]
-

p1.png (330.93 KiB) Viewed 1710 times

-
This is the title card, and the IMAGE "OPEN" contains the following script:

Code: Select all

on mouseUp
   go next
end mouseUp

[richmond62]

Here is the second card of the stack:

This contains (on the left) a reduced image of how the exercise should be completed as a set of written instructions might use words and grammatical structures with which the pupils are not yet familiar.
-

p2.png (457.06 KiB) Viewed 1709 times

-
It contains a cardScript that sets the card up to its 'virgin' state after it has been visited:

Code: Select all

on openCard
   set the vis of img "NOPE" to false
   set the vis of img "YAY" to false
   set the vis of img "arrow-R.png" to false
   put "It is           cat." into fld "xx"
end openCard

-

Yay.png (477.08 KiB) Viewed 1708 times

-

Nope.png (419.51 KiB) Viewed 1707 times

-
The images "YAY" and "NOPE" provide feedback.

[richmond62]

The "tick" image contains a script to check what a pupil has entered into the response field:
-

tick.png (28.06 KiB) Viewed 1705 times

-

Code: Select all

on mouseUp
   if fld "xx" contains "her" then
      set the vis of img "YAY" to true
      set the vis of img "arrow-R.png" to true
   else
      set the vis of img "NOPE" to true
      put "It is           cat." into fld "xx"
      wait 2 secs
      set the vis of img "NOPE" to false
   end if
end mouseUp

If the response field contains the desired response an arrow image becomes visible, allowing a pupil to proceed to the next card.

[richmond62]

As one can see, each card is graphic intensive, and, in line with what Ruth Colvin Clark

[Clark, R. 1995. Taking The Plunge. San Francisco; Macromedia, Inc.]

explained in her work, screens are uncluttered with distractions and superfluous information.

Quite a lot of the images consist of screen snaps of text blocks, or composite images consisting of text blocks overlying graphic components grouped and then imported as screen snaps:
-

p4.png (411.6 KiB) Viewed 1704 times

-

[richmond62]

NOW: in theory at least, my 2004 Master's degree thesis, using LC [at that time called 'Runtime Revolution'] proposed an agent-led interface when non-programmers (such as, for instance, ESL teachers) could answer a series on on-screen questions taking them through a decision tree to produce this sort of end product.

I have no doubt that, had that been pursued that would have been the end result: however crude and clunky the prototype now looks.
-

Screen Shot 2023-05-16 at 12.31.53 pm.png (30.18 KiB) Viewed 1703 times

-
LC claim that Xavvi can do considerably more than what my thesis stated as the end-goal of my prototype would be.

[richmond62]

My example above IS a 'real world' example, rather than the one's LiveCode central seem to have chosen to demonstrate Xavvi.

Were I LC central I would send out an appeal from 10 - 12 totally disparate real world examples and demonstrate how Xavvi could
built stacks of that sort.