[Oberon] Oberon text files

[Paul Reed]

Hi Liam, Joerg,

I can definitely identify 100% with both sides of this argument, :) in 
fact I'm reminded of F. Scott Fitzgerald's observation that "the test of 
a first-rate intelligence is the ability to hold two opposed ideas in 
mind at the same time and still retain the ability to function."[1]


>> RTF = Rich TEXT Format has text in its name but is not pure ASCII.

Actually it is - that's the point (for interchange purposes).  But what 
it *represents* is of course not plain old ASCII.


> I get it. Oberon text files are not plain text. They contain
> formatting and other info, just like RTF does.

Actually, they are binary files which are a serialisation of the 
in-memory abstract data type in Oberon called Text.  So that's why 
they're called Text, and that's why it's a bit misleading if used as an 
interchange format.

And it was never intended as an interchange format!


> *I know that they are not text.* However, it is also a fact that they
> are _called_ "text", and that word has a standard agreed meaning in
> the computer world, which it seems that Oberon does not respect.

Standard yes, agreed yes, but only when we're talking about the need to 
exchange data, ie what disparate systems need to agree on.  And although 
the original Oberon system interacted with, e.g., outside-world email 
systems, text got converted on the way in and out.

Oberon simply doesn't need to use ASCII internally, because the Text 
on-disk format is so efficient, even with formatting.  And if an ASCII 
file is accidentally opened with the Text system, it is converted to a 
Text data structure as a special case.  Various versions of the Oberon 
System do a better or a worse job of this.


> I fully understand that this is not true in Oberon but I absolutely do
> think that it is fair, reasonable and understandable to see a file
> called "README.Text" and expect it to contain plain text!

Well, strictly only if it was .txt, but I get your point and I agree 
with you in the sense that it's misleading for the beginner.

Perhaps, as a nice exercise, someone would like to write a (maybe even 
bidirectional) Oberon-Text to Markdown-subset converter, and we can 
standardise on README.md in future.

Cheers,
Paul

[1] Fitzgerald, F. Scott (1945). The Crack-Up (1st ed.). New Directions.