Opinions on CSE Validator
KarlCore
and WDG online validators?
--
Karl Core
Eric B. Bednarz
> I was wondering if anyone had any opinions on the CSE Validator
> vs. the W3C and WDG online validators?
Here's a pretty good summary:
<http://groups.google.com/groups?selm=Xns92268250213Bjkorpelacstutfi%40
193.229.0.31>
> --
<http://home.in.tum.de/~jain/software/oe-quotefix/>
might help
--
<!DOCTYPE signature [
<!ELEMENT SIGNATURE - O EMPTY>
]>
<signature>
KarlCore
"Eric B. Bednarz" <doubl...@wanadoo.nl> wrote in message
news:Xns9329C961...@130.133.1.4...
> KarlCore:
>
> > I was wondering if anyone had any opinions on the CSE Validator
> > vs. the W3C and WDG online validators?
>
> Here's a pretty good summary:
>
> <http://groups.google.com/groups?selm=Xns92268250213Bjkorpelacstutfi%40
> 193.229.0.31>
>
> > --
Cool. Its "snake oil".
Now can someone say why?
I have no plans on buying it, but someone I know was recently spouting the
merits of it, and I'd like to know why it is or is not a good tool...
--
Karl Core
David Dorward
> Cool. Its "snake oil".
> Now can someone say why?
> I have no plans on buying it, but someone I know was recently spouting the
> merits of it, and I'd like to know why it is or is not a good tool...
I've only glanced over it, but as far as I can tell:
* It encourages the use of browser specific extensions (rather then sticking
to w3 or iso standards).
* It does virtually nothing that you can't get for free in the wdg version.
So you end up forking out money for the benefit of not having to set up the
wdg validator (and then you have to spend a bunch of time tracking down and
disabling all the '<marquee> is OK' et al options.)
(As I said earlier, I didn't spend much time looking at it, so this is
rather a gut reaction)
--
David Dorward http://david.us-lot.org/
"You cannot rewrite history, not one line."
- The Doctor (Dr. Who: The Aztecs)
Eric B. Bednarz
> Eric B. Bednarz...
>> KarlCore:
>>
>> > I was wondering if anyone had any opinions on the CSE Validator
>> > vs. the W3C and WDG online validators?
>>
>> Here's a pretty good summary:
>>
>> <http://groups.google.com/groups?selm=Xns92268250213Bjkorpelacstut
>> fi%40 193.229.0.31>
> Cool. Its "snake oil".
> Now can someone say why?
It's advertised and sold as a validator while it clearly *does not and
cannot* validate. It declares pages with serious errors to be 'valid'
and valid pages to have serious 'errors'.
It's a lint, with the usual features, one of them being sharing its
author's misunderstanding and/or ignorance of specifications with the
world.
--
^^
-----------------oOo()oOo-----------------
| I saw a Volkswagen Beetle today with a |
| vanity license plate that read FEATURE |
------------------------------------------
kchayka
>
> * It encourages the use of browser specific extensions (rather then sticking
> to w3 or iso standards).
>
> disabling all the '<marquee> is OK' et al options.)
I test drove CSE Validator about a year ago. All of those
browser-specific extensions are enabled by default. They are so
entwined with standard elements and attributes that it's virtually
impossible to completely disable all of them. I never was successful at
getting validation comparable to W3C or WDG.
BTW, I asked the CSE folks why this was so. Their opinion, at the time,
was that validating just to W3C standards wasn't "real world".
--
To email a reply, remove (dash)ns(dash). Mail sent to the ns
address is automatically deleted and will not be read.
C. A. Upsdell
news:b35olv$l3$1...@ngspool-d02.news.aol.com...
> I was wondering if anyone had any opinions on the CSE Validator vs. the
W3C
> and WDG online validators?
As others have pointed out, the CSE product is not a validator. That does
not make it useless however, as AFAIK it does some useful things that a
validator does not do, e.g. check for broken links and missing anchors.
What I do is validate first (using aRealValidator.com, highly recommended),
then use a syntax checker to check for such things as broken links etc.
Headless
>As others have pointed out, the CSE product is not a validator. That does
>not make it useless however, as AFAIK it does some useful things that a
>validator does not do, e.g. check for broken links and missing anchors.
>
>What I do is validate first (using aRealValidator.com, highly recommended),
>then use a syntax checker to check for such things as broken links etc.
There are better link checkers than CSE imo, e.g.
http://home.snafu.de/tilman/xenulink.html (freeware)
I started out using CSE, the nice thing about it is that it gives much
more verbose and helpful error messages than e.g. the w3c or the ARV
validator. CSE can also be "integrated" into some authoring software.
The downside is that it flags non errors as errors while at the same
time it fails to spot some genuine errors. Having taken CSE's "advice" I
fixed the "errors" it reported, later after I switched to a proper
validator and after getting proper advice I had to go back to undo those
"fixes" again.
Despite the more verbose error messages, I'd recommend skipping over CSE
and starting with a proper validator.
Headless
Albert Wiersch
Hi Karl,
I'm the author of CSE HTML Validator.
As others have pointed out, it is not a formal validator. However, it is
quite useful as it finds many things that a formal validator can't. It also
checks CSS, links, and spelling. Please see
http://www.htmlvalidator.com/htmlval/whycseisbetter.html
Be sure that the opinions you consider about CSE are from people who have
actually USED the software (and not from those who just don't like the "HTML
Validator" name because CSE is not a formal validator--hence the "snake oil"
comment). If you talk to those who have used it, you'll likely get comments
like these (actual testimonials from real people)
http://www.htmlvalidator.com/htmlval/testimonials.html
But you are welcome to decide for yourself by downloading the trial version.
--
Albert Wiersch
AI Internet Solutions
sup...@htmlvalidator.com
http://www.htmlvalidator.com/
"KarlCore" <eightni...@SPELLOUT-893.com> wrote in message
news:b35olv$l3$1...@ngspool-d02.news.aol.com...
Jukka K. Korpela
> Hi Karl,
This reminds me of the general observation "If a message begins with
addressing an individual, it was probably not worth sending by E-mail
either." (For more of these, see
http://www.cs.tut.fi/~jkorpela/usenet/laws.html )
> I'm the author of CSE HTML Validator.
Thanks for you honesty, this time. So you are defending your own product.
You are known on Usenet as a proponent of that product, hardly otherwise.
This may help people in putting things into perspective.
> As others have pointed out, it is not a formal validator.
It is not a validator, and you _still_ keep calling it a validator, despite
the several occasions where this has been explained.
Either it is a validator, or it is not. Adding a nonsensical attribute like
"formal" won't save you. Validation is a formally defined task.
Your excuse is like calling a horse a cow and defending yourself by saying
"it is not a bovine cow, but ...".
> However, it
> is quite useful as it finds many things that a formal validator can't.
It surely finds "many things".
But you don't say what it actually does. Apparently because part of the
"validation" it performs is just invented tests. _You_ declare something as
"errors" and sell a product to report them. It just adds to the confusion
that _some_ of the messages actually report errors or justified warnings.
How is the user expected to know which is which?
> Be sure that the opinions you consider about CSE are from people who
> have actually USED the software
Well, it should be enough to anyone with a clue to try your free sample:
typing http://www.w3.org/ into the text input box at your
http://www.htmlvalidator.com/ and clicking on "Check URL" gives us:
"CSE HTML Validator Professional v5.51 generated 3 error messages, 1
warning message, and 17 other messages when checking this web page. These
problems may damage this web page's search engine rankings as well as cause
viewing problems for visitors. It is highly recommended that any problems
be corrected."
Now that I have used your "validator" once again, I know it's bogus. You're
not telling what "errors" your "validator" finds on the W3C main page, and
this is part of the snake oil selling, and quite sufficient evidence.
--
Yucca, http://www.cs.tut.fi/~jkorpela/
Pages about Web authoring: http://www.cs.tut.fi/~jkorpela/www.html
Albert Wiersch
Hi Jukka,
I'm not going to argue over the name again. If you don't like it, that's
your option.
Below are the messages that CSE generates for the W3C web page. You can
ignore #5 if you'd like since you are using it properly, but CSE alerts you
to when you might not be. As for message 2 and 3, why does label not have an
end tag? The XHTML Recommendation says "All elements [in HTML 4] other than
those declared in the DTD as EMPTY must have an end tag". If I am reading
this correctly, since "label" is not an empty tag in HTML 4, it must have an
end tag. Why doesn't the W3C page use an end tag for "label" in accordance
with the XHTML recommendation? CSE also displays message #4 which the W3C
doesn't catch because technically it is not an error but in reality it
doesn't make any sense (and that's why CSE finds it).
Validator message export for "http://www.w3.org/":
Generated by CSE HTML Validator Professional v5.51
(http://www.htmlvalidator.com/)
1. Warning in line 224 at character 5: [43] You've used the "style"
attribute but have not defined a default style sheet language (note that
HTML Validator would not see this declaration if it is sent as an HTTP
header by a web server). HTML 4.01 and XHTML require this for valid
documents. For example, include this in the "head" section of your document
to specify "text/css" as the default style sheet language: <meta
http-equiv="Content-Style-Type" content="text/css" />. See
http://www.w3.org/TR/html4/present/styles.html#default-style If you are
using HTML Validator's integrated editor, then you can add this from the
Tags menu and/or from the Tag Inserter.
2. Error in line 350 at character 5: [109] The "label" tag should be closed
by using an end tag (like </label>). It should not be closed by adding a
slash to the end of the tag (<label ... /> is not recommended). Note that
this may be technically valid according to XML parsers and validators but it
may not be technically valid according to the XHTML specification because
all elements that are not declared as EMPTY must have an end tag. In any
case, it is highly recommended that you use an end tag for backward
compatibility. Even newer browser such as Internet Explorer 6.0 may have
problems displaying your page correctly if you do not use a separate closing
tag.
3. Error in line 352 at character 5: [109] The "label" tag should be closed
by using an end tag (like </label>). It should not be closed by adding a
slash to the end of the tag (<label ... /> is not recommended). Note that
this may be technically valid according to XML parsers and validators but it
may not be technically valid according to the XHTML specification because
all elements that are not declared as EMPTY must have an end tag. In any
case, it is highly recommended that you use an end tag for backward
compatibility. Even newer browser such as Internet Explorer 6.0 may have
problems displaying your page correctly if you do not use a separate closing
tag.
4. Error in line 356 at character 7: You should only use the "checked"
attribute with the "input" element when "type=checkbox" or "type=radio".
5. Warning in line 452 at character 19: [75] The "href" attribute specifies
an invalid internal link. An internal link name should normally follow the #
character. However, if you are referencing the top of a document, then this
is valid and this message should be ignored.
--
Albert Wiersch
AI Internet Solutions
sup...@htmlvalidator.com
http://www.htmlvalidator.com/
"Jukka K. Korpela" <jkor...@cs.tut.fi> wrote in message
news:Xns932C6144D66B...@193.229.0.31...
Alan J. Flavell
> Hi Jukka,
You're addressing usenet. If you want a one to one dialogue, there's
email.
> I'm not going to argue over the name again.
There's nothing to argue about. It says 'validator', in an HTML
context. The term 'validator' has a precise meaning in an HTML
context. It isn't one. End of story.
> If you don't like it,
It isn't about "liking", it's about factual accuracy. It might be a
fine lint or checker, if it had been offered as such (though I wasn't
particularly happy with it the last time I looked).
And if you're going to post to usenet, how about a soupcon of
sensitivity for Usenet netiquette? (not that I'm holding my breath).
[TOFU environmentally recycled]
Nick Kew
at the keyboard of "Albert Wiersch" <intern...@wiersch.com> wrote:
> Hi Jukka,
>
> I'm not going to argue over the name again. If you don't like it, that's
> your option.
It's also a lie. That's what upsets people who care about honesty.
It's also a source of confusion, and as such is actively damaging.
> Below are the messages that CSE generates for the W3C web page. You can
Indeed, you have some reasonable warnings in there. Other tools such
as AccessValet or Tidy would doubtless generate a few warnings too.
The key difference is that some tools are honest about what they're
doing.
> ignore #5 if you'd like since you are using it properly, but CSE alerts you
> to when you might not be. As for message 2 and 3, why does label not have an
> end tag?
It's useless (and therefore IMO fair to warn about), but how is it
either harmful or invalid?
> with the XHTML recommendation? CSE also displays message #4 which the W3C
Same comment.
Do you think I'd get more sales if I advertised AccessValet as a
validator? I may be at a disadvantage here, because I'm not going
to make any such claims for it until and unless a future version
fully supports validation.
--
Nick Kew
Available for contract work - Programming, Unix, Networking, Markup, etc.
Albert Wiersch
"Nick Kew" <ni...@fenris.webthing.com> wrote in message
news:otob3b...@jarl.webthing.com...
>
> Indeed, you have some reasonable warnings in there. Other tools such
> as AccessValet or Tidy would doubtless generate a few warnings too.
> The key difference is that some tools are honest about what they're
> doing.
I never claimed CSE to be a formal validator.
>
> > ignore #5 if you'd like since you are using it properly, but CSE alerts
you
> > to when you might not be. As for message 2 and 3, why does label not
have an
> > end tag?
>
> It's useless (and therefore IMO fair to warn about), but how is it
> either harmful or invalid?
It shouldn't be there even if it is harmless or valid. If an author doesn't
care about this, they can simply tell CSE not to generate that message. In
my opinion, a good author would care about it because it is sloppy to
generate such markup. And sloppy markup can be inefficient... so you could
say that it is harmful because it is inefficient markup.
>
> > with the XHTML recommendation? CSE also displays message #4 which the
W3C
>
> Do you think I'd get more sales if I advertised AccessValet as a
> validator? I may be at a disadvantage here, because I'm not going
> to make any such claims for it until and unless a future version
> fully supports validation.
I don't call it a validator to dishonestly generate more sales. I don't
claim that it is a formal validator. It's an informal validator. I realize
some people disagree, but it's their option to.
Albert
Nick Kew
> claim that it is a formal validator. It's an informal validator. I realize
On this newsgroup that's harmless - there are enough people here to correct
any false impression that might give.
Elsewhere it clearly *does* give a misleading impression. We far too often
have to clear up confusion by explaining what validation is (and isn't).
I once saw an advert from an american company selling degrees "from" a
"University of Oxford". They presumably had weasel-words to justify
that, too.
Eric B. Bednarz
> "Albert Wiersch" <intern...@wiersch.com> wrote:
>
>> I don't call it a validator to dishonestly generate more sales. I
>> don't claim that it is a formal validator. It's an informal
>> validator.
Oh dear.
AFAICS nobody in this threat^Hd called you formally (eg in a court of
law) dishonest; there's just been an informal notification of utter
untruthfulness concerning the advertised possibilities of the
application in question.
Coincidentally, my killfile isn't a formal mechanism to fight idoacy,
it's just an informal way of adjusting my personal experience of signal
to noise ratio on abusenet.
> On this newsgroup that's harmless - there are enough people here
> to correct any false impression that might give.
>
> Elsewhere it clearly *does* give a misleading impression.
Hitting the nail entirely on its head.
"I found this cool tag^H^H^H'validator' to fix my site."
*mumble*
Daniel R. Tobias
> I was wondering if anyone had any opinions on the CSE Validator vs. the W3C
> and WDG online validators?
My own article on validators, linters, and the difference between the two:
http://webtips.dan.info/validators.html
--
== Dan ==
Dan's Web Tips: http://webtips.dan.info/
Dan's Domain Site: http://domains.dan.info/
C. A. Upsdell
news:j68h5vkf5sumbrq93...@4ax.com...
> "C. A. Upsdell" <cupsdell@-@-@upsdell.com> wrote:
>
> >As others have pointed out, the CSE product is not a validator. That
does
> >not make it useless however, as AFAIK it does some useful things that a
> >validator does not do, e.g. check for broken links and missing anchors.
> >
> >What I do is validate first (using aRealValidator.com, highly
recommended),
> >then use a syntax checker to check for such things as broken links etc.
>
> There are better link checkers than CSE imo, e.g.
> http://home.snafu.de/tilman/xenulink.html (freeware)
I used link checking as an example of something that can be done using a
lint program that is not done by a validator. XenuLink is good -- I use it
for checking for broken external links -- but it does much less than a good
lint program.
kchayka
>
> I never claimed CSE to be a formal validator.
IEEE Standard Glossary of Software Engineering Terminology, IEEE Std
610.12-1999:
VALIDATION
"The process of evaluating a system or component during or at the end of
the development process to determine whether it satisfies specified
requirements."
CSE does not validate. An HTML validator would mark as pass/fail
against the specified DTD. CSE doesn't even use a DTD, so how can it
validate?
Chuck Taylor
wrote:
>CSE does not validate. An HTML validator would mark as pass/fail
>against the specified DTD. CSE doesn't even use a DTD, so how can it
>validate?
Already covered earlier. The word "validator" doesn't really mean
"validator" until you put the word "formal" in front of it.
--
Chuck Taylor
President of the United States (Informal)
Albert Wiersch
"kchayka" <kcha-...@sihope.com> wrote in message
news:3e5a5e55$1...@news.sihope.com...
> Albert Wiersch wrote:
> >
> > I never claimed CSE to be a formal validator.
>
> IEEE Standard Glossary of Software Engineering Terminology, IEEE Std
> 610.12-1999:
>
> VALIDATION
> "The process of evaluating a system or component during or at the end of
> the development process to determine whether it satisfies specified
> requirements."
>
> CSE does not validate. An HTML validator would mark as pass/fail
> against the specified DTD. CSE doesn't even use a DTD, so how can it
> validate?
It doesn't validate according to that definition IF the requirement is to
validate EXACTLY against the HTML DTD. Like I've said before, I never
claimed it to. It is not a formal validator.
If the requirement of the user is to check for a combination of HTML
standards (but not strict DTD compliance), what works in the real word, and
to catch problems that a DTD validator cannot find (very helpful, I might
add!), then it would fit that definition of validator.
It validates according to this definition of validator: "to corroborate the
soundness of". I believe that to be the common usage definition of the word.
Maybe not in this newsgroup though.
Albert
richard
chuck.taylor@spamtrap says...
> On Mon, 24 Feb 2003 11:58:41 -0600, kchayka <kcha-...@sihope.com>
> wrote:
>
> >CSE does not validate. An HTML validator would mark as pass/fail
> >against the specified DTD. CSE doesn't even use a DTD, so how can it
> >validate?
>
>
>
> Already covered earlier. The word "validator" doesn't really mean
> "validator" until you put the word "formal" in front of it.
>
Isofarro
>
> "kchayka" <kcha-...@sihope.com> wrote in message
> news:3e5a5e55$1...@news.sihope.com...
>> CSE does not validate. An HTML validator would mark as pass/fail
>> against the specified DTD. CSE doesn't even use a DTD, so how can it
>> validate?
>
> It doesn't validate according to that definition IF the requirement is to
> validate EXACTLY against the HTML DTD.
So its not a validator, yet it calls itself a validator. Seems awfully
fradulent to me.
--
Iso.
FAQs: http://html-faq.com http://alt-html.org http://allmyfaqs.com/
Recommended Hosting: http://www.affordablehost.com/
Web Standards: http://www.webstandards.org/