MobileRead Forums - View Single Post - What "Cleaning Up" Do Project Gutenberg Texts Need [closed]

GregS · 11-09-2007, 07:42 AM

PPS

Is this the ZML?

http://rx4rdf.liminalzone.org/ZMLMarkupRules

If it is I have some preliminary opinions.

"Like the Wiki and SLiP formats its goal is to be a human-friendly markup language: simple, clear, and concise."

What I like about the bulkiness of XML is that it is passive and robust, and simple to repair - how simple, clear and concise it is from this point of view largely irrelevant. Beginning and end tags are bulky and inefficient, but they make things robust. A typo, an accidental deletion etc., can leave clues behind - more efficient marking up also means more fragile.

But no one in their right mind wants to directly work with XML tags ( or should not, I suspect that many that do are not in their right minds, or just trapped by current technology).

As a processing and composing language ZML has some virtues, but for me this is not enough, things have to be future proofed and robustly constructed - that means that redundancies (such as element tags for closure) that are in fact a good thing for storage, stablity, flexiblity and preservation, in terms of rendering transmission etc., the overhead of preprocessing XML into another form is well worth the costs.

ZML as a method no doubt has its uses.

Two other approaches I prefer, REBOL and LUA read XML directly into their data structures, which can then be manipulated by the script language like any other data. Both can save out as XML at any time.

I don't know if this makes sense in this context, but I favour lean applications and fat data. I have been with computers since the Apple II, from when everything was a squeeze. I don't really miss the old applications, but the loss of data still hurts, i have long thrown out a lot of things (half-written books, notes, articles etc.,.) I have written simply could not be read.

Digital Preservation is a critically important issue. The solution is robust redundancies, rich data, and most of all simplicity. The cost is fatter files, a little extra processing overhead - and it is all well worth the price, if we can preserve unambiguously what we already have.

Writing on paper disappeared only with the paper itself rotted away. No new improvements in the press, bindings or reading glasses effected what had been preserved.

What was written before did not disappear when a new pair of glasses were purchased. Until recently new digital glasses made previous words disappear. XML and XHML preserve some aspects well, but other standards are needed to preserve it better (TEI in the end I suggest for Literature).

ZML has its place, but not as a method of storage.

11-09-2007, 07:42 AM	#131
GregS Zealot Posts: 107 Karma: 308 Join Date: Oct 2007 Location: Perth Australia Device: EZ Reader 5", Iliad	PPS Is this the ZML? http://rx4rdf.liminalzone.org/ZMLMarkupRules If it is I have some preliminary opinions. "Like the Wiki and SLiP formats its goal is to be a human-friendly markup language: simple, clear, and concise." What I like about the bulkiness of XML is that it is passive and robust, and simple to repair - how simple, clear and concise it is from this point of view largely irrelevant. Beginning and end tags are bulky and inefficient, but they make things robust. A typo, an accidental deletion etc., can leave clues behind - more efficient marking up also means more fragile. But no one in their right mind wants to directly work with XML tags ( or should not, I suspect that many that do are not in their right minds, or just trapped by current technology). As a processing and composing language ZML has some virtues, but for me this is not enough, things have to be future proofed and robustly constructed - that means that redundancies (such as element tags for closure) that are in fact a good thing for storage, stablity, flexiblity and preservation, in terms of rendering transmission etc., the overhead of preprocessing XML into another form is well worth the costs. ZML as a method no doubt has its uses. Two other approaches I prefer, REBOL and LUA read XML directly into their data structures, which can then be manipulated by the script language like any other data. Both can save out as XML at any time. I don't know if this makes sense in this context, but I favour lean applications and fat data. I have been with computers since the Apple II, from when everything was a squeeze. I don't really miss the old applications, but the loss of data still hurts, i have long thrown out a lot of things (half-written books, notes, articles etc.,.) I have written simply could not be read. Digital Preservation is a critically important issue. The solution is robust redundancies, rich data, and most of all simplicity. The cost is fatter files, a little extra processing overhead - and it is all well worth the price, if we can preserve unambiguously what we already have. Writing on paper disappeared only with the paper itself rotted away. No new improvements in the press, bindings or reading glasses effected what had been preserved. What was written before did not disappear when a new pair of glasses were purchased. Until recently new digital glasses made previous words disappear. XML and XHML preserve some aspects well, but other standards are needed to preserve it better (TEI in the end I suggest for Literature). ZML has its place, but not as a method of storage.