<FAQENTRY> 3.2 update needed

[Dr John Stockton]

JRS: In article <[email protected]>, dated Tue, 1
Nov 2005 19:17:47, seen in Richard Cornford

As to the FAQ being a wiki; I have said from the outset, and it is
stated in the notes, that anyone wishing to contribute an article to the
FAQ notes is free to do so, so long as they accept that the article be
subject to public technical scrutiny on the group (and accepted as
accurate/useful) and may be subject to unrestricted future editing (in
the event that it becomes inaccurate at some future point).

If an article is worth putting in the FAQ notes, then it is worth
posting here, initially and after any *major* change. It would be
sufficient to copy'n'paste from a browser view.

The total number of articles proposed for inclusion in the FAQ notes in
the last two years is one. And that written by VK; factually incorrect,
confused and misleading, and so not included because it did not (and
could not) pass the scrutiny of the group.

Your notes are not conveniently accessible to me, since I dial up only
briefly.

They are not collectively indexed in the FAQ, and I don't know if my own
index is up-to-date. I suggest that they be indexed in a [sub-]section
of the FAQ, with modification date; and that there should be a Web index
with modiification dates (you may already have one) specifically
mentioned in the FAQ.

And we are not even restricted to whole articles, there is the
miscellaneous tips and tricks page, where any well explained specific
technique might be presented, but no contributions there either.

So given that nobody seems willing or able to contribute anything
substantial to the FAQ

As you don't update the posted FAQ frequently, you cannot expect many
people to bother to prepare suggestions.

 

[Dr John Stockton]

JRS: In article <[email protected]>, dated
Wed, 2 Nov 2005 08:54:41, seen in Paul Cooper

Translation is very useful, of course, but people need to be aware
which version of the FAQ has been translated, as this is a volunteer
activity that will depend on the time available.

Put at the top of each smallest-translateable-unit the date of its last
translation-affecting change -- that will be useful for those who only
read the English too -- and in the translation, copy that and perhaps
add the date of translation.

 

[Dr John Stockton]

JRS: In article <[email protected]>, dated Tue, 1 Nov
2005 19:02:47, seen in Randy Webb

Dr John Stockton said the following on 11/1/2005 4:27 PM:

True, but, I don't see why the FAQ Editor has to be a "knowledgeable
person" if that knowledge is in reference to scripting itself. Other
than knowing how the server side processing works. It only takes a
person that can review, count votes, and edit text to maintain it. The
auto-posting of the FAQ is done automatically.

An editor without a good knowledge of both the languages - English and
Javascript - cannot be relied upon to produce a comprehensible and exact
version. An editor who has to be continually corrected in detail will
not last, and meanwhile wastes everyone's time.

However, if we are talking about a mere publisher, it would be possible
for someone knowing no javascript and being able to read but not write
English to transfer sections according to agreed instructions here; I
doubt whether anyone would want to do that for long unpaid.

 

[Dr John Stockton]

JRS: In article <[email protected]>, dated Wed, 2
Nov 2005 00:29:27, seen in RobG

Has anyone looked at wikipedia lately? It's free, supports mulitple
languages and already has a lot of JavaScript stuff - maybe a bit of
effort from those here will keep it up-to-date?

It may be a useful adjunct to the jibbering.com site, though not really
a replacement.

Here's an example of why, to be reliable, a FAQ needs a well-chosen
Editor combined with review by others : the site containing http://www
..criosoftware.it/default42.aspx seems frequently cited in
(though AFAICS only by one person), and
that page contains
function __pt_str_codXcifre(nchar,car) {
var str="";
for(var x=0;x<eval(parseInt(nchar)-this.length);x++) str+=car;
return str+this;
}
In use, 'this' is String. Note eval and parseInt(x), in a for loop
which should be a while loop. While not involving the content of str in
every concatenation is good, ISTM that the benefit may be outweighed by
the cost of the extra concatenation to add str.

That's the sort of code that might well be put in a wiki-faq, perhaps
even as a replacement for something better. One cannot safely cite a
wiki if one wishes to cite a particular solution.


The page also contains "Inserita il 02/11/2005" which should mean
yesterday but may mean February.



<FAQENTRY> in 4.6, change 0.07 to 0.06+0.01 ???

FAQ has 4.7 Why does 5 * 1.015 != 5.075 or 0.05+0.01 != 0.06?
Better? 4.7 Why does 5 * 1.015 != 5.075 or 0.06+0.01 != 0.07?





Note that we normally adjust length by incrementing, which in javascript
requires a new string every time.

For generating long strings, the following recursive routine takes many
fewer operations; there must be a way to code it iteratively for
efficiency. Lightly-tested; but derived from something reliable for
generating positive integer powers in a language without a Math.pow
equivalent. Using L<0 is not recommended.

function BigCat(L) { var T
if (L==0) return ""
if (L&&1) return BigCat(L-1) + " "
T = BigCat(L/2) ; return T + T }

 

[Matt Kruse]

<snip bad code>
That's the sort of code that might well be put in a wiki-faq, perhaps
even as a replacement for something better. One cannot safely cite a
wiki if one wishes to cite a particular solution.

If it replaces something better, then surely it will be rolled back by
someone else who notices the bad change.
Have you used wikipedia much?
The quantity and quality of information there is astounding, even though
anyone can edit anything.

If a wiki javascript faq were monitored by a number of knowledgeable people,
bad changes would not survive.

Unless you're volunteering to keep the FAQ more relevant and up-to-date, I
don't see anyone volunteering to be the single point of contact for what has
become a huge document.

 

[VK]

If it replaces something better, then surely it will be rolled back by
someone else who notices the bad change.
Have you used wikipedia much?
The quantity and quality of information there is astounding, even though
anyone can edit anything.

If a wiki javascript faq were monitored by a number of knowledgeable people,
bad changes would not survive.

Unless you're volunteering to keep the FAQ more relevant and up-to-date, I
don't see anyone volunteering to be the single point of contact for what has
become a huge document.

What I'm not sure about "wiki" is the inertial mass of the system.

Say some ill-minded person will edit an array-related faq by saying
that array exists as separate unity, it has length property and that
property indicates to the total amount of elements in the array ?
Also anyone who's capable to pronounce such terrible things can also
add a link at the bottom leading to some hard porno site or IE download
site.
So that guarantees do we have that this article will return in the
correct state in some period of time? How many visitors will read the
article in its violated state. And if no one will take time to edit
this article at all? I'm not sure how does it work in the wikipedia.com

 

[Randy Webb]

VK said the following on 11/4/2005 5:06 AM:

What I'm not sure about "wiki" is the inertial mass of the system.

Say some ill-minded person will edit an array-related faq by saying
that array exists as separate unity, it has length property and that
property indicates to the total amount of elements in the array ?

That happens almost weekly here, with people saying they have created
Hash Arrays in Javascript, but you have a point.

Also anyone who's capable to pronounce such terrible things can also
add a link at the bottom leading to some hard porno site or IE download
site.

That can happen, easier, in Usenet. And the results get propogated wider
and quicker. The advantage of a Wiki is that it *can* be corrected, and
the incorrect data removed. Unfortunately, in Usenet, once you have said
"I have created a Hash Array in Javascript", it can only be corrected,
not removed.

So that guarantees do we have that this article will return in the
correct state in some period of time? How many visitors will read the
article in its violated state. And if no one will take time to edit
this article at all? I'm not sure how does it work in the wikipedia.com

Go to wikipedia.org and read up on it.

 

[Dr John Stockton]

JRS: In article <[email protected]>, dated Thu, 3 Nov 2005
20:04:30, seen in Matt Kruse

If it replaces something better, then surely it will be rolled back by
someone else who notices the bad change.

Perhaps, eventually. But, while one can expect the experts here to
review each new edition of *our* FAQ soon after it appears, one can
hardly expect them to with all relevant wiki pages continually.

Have you used wikipedia much?
The quantity and quality of information there is astounding, even though
anyone can edit anything.

Most articles I find somewhat shallow - though adequate for what the
average user wants, no doubt - printed encyclopaedias are better, being
written by carefully-chosen experts rather than self-appointed ones. I
have found wiki to be in error, too.

There is or was an error in the Zeller part of de.wiki (main wiki did
not have the part where the error was), and ISTR an error about Germany
in the DST part of the main wiki - can we assume they've been corrected
by now? The true facts are elsewhere, in Risks Digest and in an obvious
place on the Web.

If a wiki javascript faq were monitored by a number of knowledgeable people,
bad changes would not survive.

If and only if.

Unless you're volunteering to keep the FAQ more relevant and up-to-date, I
don't see anyone volunteering to be the single point of contact for what has
become a huge document.

The FAQ is not particularly large; Select All on the 2004 HTML issue
gets just under 30,000 characters. For Y2k, I maintained a UK FAQ which
ended up at just under 31,000 characters.

However, I used a simpler system, plain text, and did not allow
suggestions to build up. I put it on my Web site, uploading all changes
at next dial-up; AW kindly CRONned it from there into News early each
Sunday.


I have already suggested that Jim could amend the (presumed) Mon/Fri
CRON job to fire only on Mondays, and start a new CRON job for Fridays -
or /vice versa/. The new job would post updates to the older material,
as a plain text file comprising chunks copied from News.

That would in essence be a wiki, but with moderated update.

I write a small section and post it to News; it's copied in if Mod likes
it. You improve it; Mod likes that better, and copies it over mine. Ej
suggests a good sentence; Mod copies that in. PE posts nonsense in a
changed version; Mod ignores it. LN does a total rewrite; Mod puts that
in as a replacement.

Once a section is stable and goof, then the FAQ editor can extract it,
rephrase it convert it to XML or whatever, and drop it in the Monday
FAQ.