Cite Sources! Dispell FUD!
Dan Connolly
It's been a long time since I visited USENET newsgroups about HTML and the web.
I'm encouraged to see healthy debate of HTML specifications, standards,
and extensions along with the general HTML techniques and how-to dialogue.
But I'm concerned with the level of Fear, Uncertainty, and Doubt in
many of these postings. What's the authoritative source of info on
HTML? Which browsers support what? Which browsers are broken in what
way?
I don't have anything specific to add to the debate. You can find
my opinions elsewhere. But I have a general recommendation:
An answer posted to USENET without references isn't worth the
bandwidth used to transmit it.
If you're answering an HTML syntax question, cite your source.
There's really no excuse not to. The beauty of internet technology is
that anybody can get at the authoritative specs in a matter of
seconds. And the beauty of the web is that you can make and follow
these references so easily.
Give a URL. And since web documents can change, give the revision
date. And since web documents can move, give the title and author, so
folks can use search services to find moved documents. (Crediting the
author is really a matter of politeness, too.) If you want to
save somebody a little time, excerpt the relevant portion of
the document.
Your argument will be so much more convincing if the reader can check
your answer from independent sources. You'll build
credibility. Quality of documents on the web will increase! World
hunger will be eliminated!!!! Er... ahem...
Dan
--
Daniel W. Connolly "We believe in the interconnectedness of all things"
Research Scientist, MIT/W3C PGP: EDF8 A8E4 F3BB 0F3C FD1B 7BE0 716C FF21
<conn...@w3.org> http://www.w3.org/hypertext/WWW/People/Connolly
Phil Stripling
>SNIP<
: An answer posted to USENET without references isn't worth the
: bandwidth used to transmit it.
On what do you base this opinion?
>SNIP<
: Your argument will be so much more convincing if the reader can check
: your answer from independent sources. You'll build
Why do you think this is so?
: credibility. Quality of documents on the web will increase! World
I disagree.
>SNIP<
--
Phil
PGP+ M+ E W++(+) U P+ w--
Frank McNeil
: Dan Connolly (conn...@w3.org) wrote:
: >SNIP<
: : An answer posted to USENET without references isn't worth the
: : bandwidth used to transmit it.
: On what do you base this opinion?
In my proffesional experience; its just common sense. In general
only real experts can give authoritative answers; the rest is
conjecture. Most real experts don't have the time to browse low
signal to noise ratio newsgroups; hence it is helpful that people
paste in the references they used to come to their conclusions. The
business buzzword is "confirmable reference."
I remember I used to say over and over when referring to income tax
issues, that an answer without references was BS. I found that people
really got sidetracked when they discussed things and did not
consistently refer to authoritative references. IMO, issues get
resolved faster if people use good reasoning with authoritative
references. Actually things are also a lot simpler if we don't get
sidetracked by opinions we have to disprove later.
An answer without a reference is usually only a question for serious
people until they get the reference. Well that's my experience.
: >SNIP<
: : Your argument will be so much more convincing if the reader can check
: : your answer from independent sources. You'll build
: Why do you think this is so?
I'll speak for myself; when I say that there is no way to know if the
author of a statement has used enough analysis and sound references to
prove that his or her answers are correct or reasonable.
By the way, there are a few experts in this group; note that many of
them always provide references. It's just the normal thing that
serious people do.
: : credibility. Quality of documents on the web will increase! World
: I disagree.
Stange attitude; does it have something to do with fuzzy logic?
frank
Rich...@stars.com
>
> It's been a long time since I visited USENET newsgroups about HTML and the
web.
I heartily endorse what you've said here. But I think the practical
difficulty for most people will be the tedium of the mechanics in searching
out and pulling up the relevant source and cutting and pasting it into their
article. This might not be too much of a burden for us veterans (except for
our advanced senility..) but I doubt most of the respondents to these
newsgroups - for better or worse - will do as you wisely suggest.
However, what might help is some kind of compromise: I know for
example that the authoritative source for HTML 3.0 is Dave Raggett's Internet
Draft at http://www.w3.org/hypertext/WWW/MarkUp/html3/CoverPage.html
It took me several mouse clicks to get there, along a well-travelled path from
the WDVL/Specifications page [WDVL/Specs] to the W3C, and I knew exactly where
I was going. I might have traded keystrokes for mouse clicks but still, it was
tedious - because the end result is nothing novel. Well, we don't have URNs
yet, but maybe we could cite such well-known sources by an abbreviation, e.g.
[W3C/HTML3.0] (with some TBD micro-syntax). Together with a well-known
"dictionary" for looking up, people might then be happy to cite their sources,
in those cases where they are "well known".
Just my $20 worth.
Alan Richmond
[WDVL/Specs] = http://WWW.Stars.com/Specs/
>
> An answer posted to USENET without references isn't worth the
> bandwidth used to transmit it.
>
Alan_Richmond____________________The Web Developer's Virtual Library
http://WWW.Stars.com/CWSW/ CyberWeb SoftWare____ USA (301)
http://WWW.Stars.com/Spectrum/ The WDVL Journal | Tel: 552 0272
http://WWW.Enterprise.net/stars/ UK mirror of the WDVL| Fax: 552 4306
Phil Stripling
: : >SNIP<
: : : An answer posted to USENET without references isn't worth the
: : : bandwidth used to transmit it.
: : On what do you base this opinion?
: In my proffesional experience; its just common sense. In general
Common sense? Do you have a reference for this? A confirmable reference?
>SNIP<
: I remember I used to say over and over when referring to income tax
: issues, that an answer without references was BS. I found that people
So what did you cite the people to? The code? TreasRegs? Could they then
go home and read it? And understand it?
: really got sidetracked when they discussed things and did not
: consistently refer to authoritative references. IMO, issues get
: resolved faster if people use good reasoning with authoritative
: references. Actually things are also a lot simpler if we don't get
: sidetracked by opinions we have to disprove later.
Do you have any reference for this?
>SNIP<
: I'll speak for myself; when I say that there is no way to know if the
: author of a statement has used enough analysis and sound references to
: prove that his or her answers are correct or reasonable.
Can you cite me something on this? It looks like opinion from here.
: By the way, there are a few experts in this group; note that many of
: them always provide references. It's just the normal thing that
: serious people do.
Have you seen abby's posts?
>SNIP<
: Stange attitude; does it have something to do with fuzzy logic?
No, just trying to apply to the posts the requirement for references
requested in the posts. I think it's recursive rather than fuzzy. Just
trying to apply the request for references to learn the authority for the
opinions stated.
For the most part, this isn't stuff that's hard to learn -- certainly not
the IRC (as amended). It's a high noise to signal groupd of beginners.
Experts create even more noise with the html.police attitude toward
people who want counters and Netscape tags.
Some things, like tax advice, are _worth_ the effort of providing
citations to authority. This group isn't worth the effort of providing
cites to the html DTD. 98% of the people reading this group (including
me) can't _read_ the DTD and make sense of it.
But then, that's just my opinion, and I have no confirmable references.