Thank you all very much indeed!

I see that there are indeed more ways of skinning a cat than choking
it to death with butter!

My latest preference for what I am trying to achieve is the following

<style type="text/css">
h2 {
padding-bottom: 0;
margin-bottom: 0;
color: blue
}

h2 + p {
text-indent: 20px;
color: red;
padding-top: 0;
margin-top:0;
}


This gives me an h2 with no bottom spacing (what I want) and a p with
no top spacing when it follows an h2, but I DO get inter p spacing
without having to add it manually.  I am pleased with this solution! 
Thanks  again and hopefully I shall be quiet now for a day or two
while I update my magnum-opus accordingly!

At 09:35:46 on Sat, 14 Mar 2009, me here  wrote in 
:

>Thanks  again and hopefully I shall be quiet now for a day or two
>while I update my magnum-opus accordingly!

The next thing you need to think about, I'm afraid, is your Usenet 
persona.  It's up to you if you don't want to give us a name;  however, 
I very much doubt whether you are actually Michael Grossman - so why are 
you using the domain here.com, which belongs to him, and exposing it to 
spam?  Why not use a truly .invalid e-mail address format in your From, 
and include a working Reply-To (which does not get auto-harvested for 
spam, because it does not form part of the subset of headers which the 
bots look at)?

As an example:  mail to my From address of 
 is automatically blackholed, whereas 
mail to my ReplyTo address of  is 
accepted - and, despite having been in use for some three years, has not 
yet started to receive spam.  (When it does, it can be easily changed 
for something else.)

If you don't want even to suggest your own e-mail address in your posts, 
even with an .invalid marker, set up a webmail account which you never 
read, and use that in your From.  Honestly, using somebody else's e-mail 
address without their permission is really *not* the done thing.

<http://en.wikipedia.org/wiki/Address_munging> may help a bit.
-- 
Molly Mockford
They that can give up essential liberty to obtain a little temporary safety
deserve neither liberty nor safety - Benjamin Franklin
(My Reply-To address *is* valid, though may not remain so for ever.)

On Sat, 14 Mar 2009 11:44:45 +0000, Molly Mockford
 wrote:

>At 09:35:46 on Sat, 14 Mar 2009, me here  wrote in 
>:
>
>>Thanks  again and hopefully I shall be quiet now for a day or two
>>while I update my magnum-opus accordingly!
>
>The next thing you need to think about, I'm afraid, is your Usenet 
>persona.  It's up to you if you don't want to give us a name;  however, 
>I very much doubt whether you are actually Michael Grossman - so why are 
>you using the domain here.com, which belongs to him, and exposing it to 
>spam?  Why not use a truly .invalid e-mail address format in your From, 
>and include a working Reply-To (which does not get auto-harvested for 
>spam, because it does not form part of the subset of headers which the 
>bots look at)?
>
<contrite>
You make a very valid point.  My error entirely on two points:

a) I didn't expect anyone actually to have the domain "here.com" with
all the eejits (me not excluded) who would use it whenever a website
impertinently demands an email address to let you do something that
doesn't need an email address - like viewing your fuel bills or
something similar!  

So, I have rather inconsiderately been using it: I shall stop. I have
a perfectly good dustbin of my own - the one for this message - I was
just lazy or in hurry and couldn't be bothered to type the extra
characters.  very very bad, really. 

 b) I do have a real "persona" in Agent - but as I hadn't sued it for
some while I forgot that it was set to what I thought was an address
that would be totally invalid.   

</contrite>

My experience of getting spam via usenet is, I'm afraid, historically
much worse than yours - and although I'm new to this HTML lark I've
big using Usenet since ... ages.   

So I should have known better and thank you for braiding me up.  I
haven't set up this "persona" properly yet, but I wanted to
acknowledge what you had said and let you know I intend to put it
right.

Lewis

"Molly Mockford"  wrote in 
message news:ZU8ZhHNti5uJFw9p@molly.mockford...
<snip>
:
: Why not use a truly .invalid e-mail address format in your 
From,
: and include a working Reply-To (which does not get 
auto-harvested for
: spam, because it does not form part of the subset of headers 
which the
: bots look at)?

Can you cite a reference for that statement please, how does 
anyone other than the bots writer/owner know what it looks for?
-- 
Wikipedia: the Internet equivalent of
Hyde Park and 'speakers corner'...
Sorry, mail to this address goes unread.
Please reply via group.

Jerry wrote:
> "Molly Mockford"  wrote in 
> message news:ZU8ZhHNti5uJFw9p@molly.mockford...
> <snip>
> :
> : Why not use a truly .invalid e-mail address format in your 
> From,
> : and include a working Reply-To (which does not get 
> auto-harvested for
> : spam, because it does not form part of the subset of headers 
> which the
> : bots look at)?

> Can you cite a reference for that statement please, how does 
> anyone other than the bots writer/owner know what it looks for?

There's probably no single reference and there's probably no one who 
really knows how the bot harvesters work in usenet, but there's a lot of 
people with a lot of experience of never getting any spam at all to 
reply-to addresses.

The one caveat to this - is that there are an explosion of websites 
which do nothing but copy usenet posts onto those sites and surround 
them with adverts.  I have no idea if those sites include reply-to 
headers on the page, if they do, then they've become fair game for web 
mail harvesters.

-- 
Tony Evans
Saving trees and wasting electrons since 1993
blog -> http://perceptionistruth.com/
olmr -> http://www.onelinemoviereviews.co.uk/
[ anything below this line wasn't written by me ]

At 13:37:25 on Sat, 14 Mar 2009, Jerry 
<mapson.scarts@btinternet.com.INVALID> wrote in 
<gpgcb8$li1$1@news.motzarella.org>:

>
>"Molly Mockford"  wrote in
>message news:ZU8ZhHNti5uJFw9p@molly.mockford...
><snip>
>:
>: Why not use a truly .invalid e-mail address format in your
>From,
>: and include a working Reply-To (which does not get
>auto-harvested for
>: spam, because it does not form part of the subset of headers
>which the
>: bots look at)?
>
>Can you cite a reference for that statement please, how does
>anyone other than the bots writer/owner know what it looks for?

It's the XOVER subset of headers which spam harvesters use, on the 
grounds of efficiency.  Reading through the entirety of every message 
would slow down their harvesting very considerably;  in order to get 
hold of Reply-To, they would have to have the body as well as all the 
headers!

You can read a bit about the News Overview Database, and the XOVER 
command which retrieves NOV information, at 
<http://en.wikipedia.org/wiki/NOV_(computers)>.  For more detail that 
you'll really want, go to RFC2822 and RFC1036.
-- 
Molly Mockford
They that can give up essential liberty to obtain a little temporary safety
deserve neither liberty nor safety - Benjamin Franklin
(My Reply-To address *is* valid, though may not remain so for ever.)

At 13:26:14 on Sat, 14 Mar 2009, spam.goes.here2@ntlworld.com wrote in 
:

>So I should have known better and thank you for braiding me up.  I
>haven't set up this "persona" properly yet, but I wanted to
>acknowledge what you had said and let you know I intend to put it
>right.

Thank you!  And I think we have, between us, probably provided the first 
known example on Usenet of such a request being made, and complied with, 
without any rancour whatsoever :-)
-- 
Molly Mockford
They that can give up essential liberty to obtain a little temporary safety
deserve neither liberty nor safety - Benjamin Franklin
(My Reply-To address *is* valid, though may not remain so for ever.)

On Sat, 14 Mar 2009 13:26:14 +0000, spam.goes.here2@ntlworld.com put
finger to keyboard and typed:
>
>a) I didn't expect anyone actually to have the domain "here.com" with
>all the eejits (me not excluded) who would use it whenever a website
>impertinently demands an email address to let you do something that
>doesn't need an email address - like viewing your fuel bills or
>something similar!  

The domains 'example.com' and 'example.net' (and similar constructions
in other TLDs) are deliberately reserved for use where you need a
real-looking domain that doesn't actually belong to anyone.

Mark
-- 
Blog: http://mark.goodge.co.uk
Stuff: http://www.good-stuff.co.uk

Say I have:

<div class="clearfix">
        <div style="float: left; width: 50%;">
        </div>
        <div style="float: right; width: 50%;">
        </div>
</div>


(the class clearfix is a technique to force that div to expand
vertically to enclose the floated elements inside it - see
<http://www.positioniseverything.net/easyclearing.html> for an example
of this.)

The floated divs are of arbitrary height.

I'd like to be able to centre an element vertically in one of the
internal divs, but I so far haven't I been able to work out a CSS-only
technique for doing so. All the solutions I've found to similar problems
rely on JS access to the DOM.

Is this going to prove to be impossible?

Daniele

In article 
<1ix16db.1vgmwnto3kupmN%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> Say I have:
> 
> <div class="clearfix">
>         <div style="float: left; width: 50%;">
>         </div>
>         <div style="float: right; width: 50%;">
>         </div>
> </div>
> 
> 
> (the class clearfix is a technique to force that div to expand
> vertically to enclose the floated elements inside it - see
> <http://www.positioniseverything.net/easyclearing.html> for an example
> of this.)
> 
> The floated divs are of arbitrary height.
> 
> I'd like to be able to centre an element vertically in one of the
> internal divs, but I so far haven't I been able to work out a CSS-only
> technique for doing so. All the solutions I've found to similar problems
> rely on JS access to the DOM.
> 
> Is this going to prove to be impossible?
> 

I suspect so. Or else some gyrations that no one would really want to 
employ. The problem is that the left float has no real way of knowing 
the height of the right float. Or if you like, the height of the parent. 
The parent sort of knows the height of the tallest child float in the 
sense that it is forced by your clearing technique or my (below) 
overflow one to grow itself to cover the tallest float. 

I assume your problem is something like this:

<http://dorayme.netweaver.com.au/verticalCentringInFloatContext.html>

I offer this in case someone is not quite sure what you want and might 
like to have a go as an exercise.

You would know of course, that an HTML table would get you the cross 
browser look you want. But not the semantics you might want. Though what 
semantics this floated DIV layout has, beats me!

But what is the realistic content of what you are doing? Is it an image 
you want to place? What is the context of the page etc. There would be 
solutions.

-- 
dorayme

In article ,
 dorayme  wrote:

> In article 
> <1ix16db.1vgmwnto3kupmN%real-not-anti-spam-address@apple-juice.co.uk>,
>  real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:
> 

> > I'd like to be able to centre an element vertically in one of the
> > internal divs, but I so far haven't I been able to work out a CSS-only
> > technique for doing so. All the solutions I've found to similar problems
> > rely on JS access to the DOM.
> > 


I assume that 

<http://dorayme.netweaver.com.au/verticalCentringTableCSS.html>

might go close to what you want, the drawback being browser support? 
Seems OK in FF, Safari, and other good modern browsers.

-- 
dorayme

dorayme  wrote:

> > > I'd like to be able to centre an element vertically in one of the
> > > internal divs, but I so far haven't I been able to work out a CSS-only
> > > technique for doing so. All the solutions I've found to similar problems
> > > rely on JS access to the DOM.
> > > 
> 
> 
> I assume that 
> 
> <http://dorayme.netweaver.com.au/verticalCentringTableCSS.html>
> 
> might go close to what you want, the drawback being browser support? 
> Seems OK in FF, Safari, and other good modern browsers.

The main problem I can see with that isn't browser support, but that
display: table-cell doesn't take margins, which are required in this
case.

Is this yours by the way:

<http://dorayme.890m.com/alt/ie8/ie8.html>

Daniele

In article 
<1ix2vst.111cquul5fahiN%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> dorayme  wrote:
> 
> > > > I'd like to be able to centre an element vertically in one of the
> > > > internal divs, but I so far haven't I been able to work out a CSS-only
> > > > technique for doing so. All the solutions I've found to similar problems
> > > > rely on JS access to the DOM.
> > > > 
> > 
> > 
> > I assume that 
> > 
> > <http://dorayme.netweaver.com.au/verticalCentringTableCSS.html>
> > 
> > might go close to what you want, the drawback being browser support? 
> > Seems OK in FF, Safari, and other good modern browsers.
> 
> The main problem I can see with that isn't browser support, but that
> display: table-cell doesn't take margins, which are required in this
> case.
> 

You can give a margin to #parent and you can give margins to the 
paragraphs. Won't these two facilities do for your purpose? 

> Is this yours by the way:
> 
> <http://dorayme.890m.com/alt/ie8/ie8.html>
> 

I hope there is no dorayme imposter... <g>

Yes, I saw this free web hosting when it came out and it looked ok for 
posting examples for usenet... Then after a while they started sticking 
javascript onto my pages and then I had trouble FTPing to it. So I gave 
up on it and have no access to remove things. I should try to contact 
the hosts, I guess... 

There are all sorts of things in the folder 'alt/' I have duplicated 
most of it elsewhere.

-- 
dorayme

dorayme  wrote:

> > > might go close to what you want, the drawback being browser support?
> > > Seems OK in FF, Safari, and other good modern browsers.
> > 
> > The main problem I can see with that isn't browser support, but that
> > display: table-cell doesn't take margins, which are required in this
> > case.
> 
> You can give a margin to #parent and you can give margins to the 
> paragraphs. Won't these two facilities do for your purpose? 

Unfortuantely, the siblings require margins between them. And they can
contain anything, not just paragraphs.

After lunch I will set up a server hosting a demo of more than the
simplified version I asked about.

> > Is this yours by the way:
> > 
> > <http://dorayme.890m.com/alt/ie8/ie8.html>
> > 
> 
> I hope there is no dorayme imposter... <g>

I spotted it at
<http://connect.microsoft.com/IE/feedback/ViewFeedback.aspx?FeedbackID=3
62541>

Dnaiele

D.M. Procida  wrote:

> > You can give a margin to #parent and you can give margins to the 
> > paragraphs. Won't these two facilities do for your purpose? 
> 
> Unfortuantely, the siblings require margins between them. And they can
> contain anything, not just paragraphs.
> 
> After lunch I will set up a server hosting a demo of more than the
> simplified version I asked about.

Here's a demo of the structures I'm using.

<http://w128.medcn.uwcm.ac.uk/medic/stripped.html>

The two-column version goes like this:

<div class="row two-cols">
        <div class="column firstcolumn">
        </div>
        <div class="column lastcolumn">
        </div>
</div>

The relevant CSS (in <http://w128.medcn.uwcm.ac.uk/styles/fixes.css>)
for the two-column version is:


div.row:after   {
                content: ".";
                display: block;
                height: 0;
                clear: both;
                visibility: hidden;
}

.two-cols .column       {width: 48.5%;}

.column {
        float: left;
        margin-left: 2.8%; 
        }

div.column.firstcolumn  {
        margin-left: 0;
        }
                
div.column.lastcolumn   {
        float: right;
        margin-left: 0 ; 
        } 

The display:table-cell version unfortunately isn't compatible with some
of the other needs, such as margins on the columns.

Daniele
-- 
Wanted: TEAC A-2300SX, Akai GX-4000D

In article 
<1ix3dj9.wq7rcmdxhpy9N%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> Here's a demo of the structures I'm using.
> 
> <http://w128.medcn.uwcm.ac.uk/medic/stripped.html>

...be nice if it was stripped to the two col prob you started with and 
just the necessary CSS for it...

-- 
dorayme

dorayme  wrote:

> > Here's a demo of the structures I'm using.
> > 
> > <http://w128.medcn.uwcm.ac.uk/medic/stripped.html>
> 
> ...be nice if it was stripped to the two col prob you started with and
> just the necessary CSS for it...

OK, have another look, I stripped it down further.

Daniele

In article 
<1ix4q0d.7h81d2lqi4t3N%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> dorayme  wrote:
> 
> > > Here's a demo of the structures I'm using.
> > > 
> > > <http://w128.medcn.uwcm.ac.uk/medic/stripped.html>
> > 
> > ...be nice if it was stripped to the two col prob you started with and
> > just the necessary CSS for it...
> 
> OK, have another look, I stripped it down further.
>
 
Is the question still how to get some element into the middle of the 
left column? I am sorry, I have been stuck on quite what is being 
discussed here. I see no candidate that you want to do *this* for in 
your example. There seems no attempt to do such a thing? Is there some 
particular element within one or other of the columns that you want to 
vertically centre?

Here is some divititis, since you are not concerned with browser support 
for table CSS, that centres an element of my choosing within Column 1.

<http://dorayme.netweaver.com.au/proc2.html>

I would not want to be doing this for real... usually in real 
situations, one can find better and more semantic markup. 

(btw I notice you use XHTML doctype, is there any particular reason? 
There are a few (minor) validation issues with your above URL)

-- 
dorayme

dorayme  wrote:

> In article 
> <1ix4q0d.7h81d2lqi4t3N%real-not-anti-spam-address@apple-juice.co.uk>,
>  real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:
> 
> > dorayme  wrote:
> > 
> > > > Here's a demo of the structures I'm using.
> > > > 
> > > > <http://w128.medcn.uwcm.ac.uk/medic/stripped.html>
> > > 
> > > ...be nice if it was stripped to the two col prob you started with and
> > > just the necessary CSS for it...
> > 
> > OK, have another look, I stripped it down further.
> >
>  
> Is the question still how to get some element into the middle of the 
> left column? I am sorry, I have been stuck on quite what is being 
> discussed here. I see no candidate that you want to do *this* for in 
> your example. There seems no attempt to do such a thing? Is there some
> particular element within one or other of the columns that you want to
> vertically centre?

The question is, given this basic structure - of sibling divs within a
container divs, to produce columns - is it possible to centre elements
vertically within one of those sibling divs, without resorting to
JavaScript?

I think the answer is that is simply not possible, or at least, not
possible without giving up control over pther aspects of the box model.

> Here is some divititis, since you are not concerned with browser support
> for table CSS, that centres an element of my choosing within Column 1.
> 
> <http://dorayme.netweaver.com.au/proc2.html>

Sadly, display:table-cell means losing control over widths and margins.

> (btw I notice you use XHTML doctype, is there any particular reason? 

Not really.

> There are a few (minor) validation issues with your above URL)

How odd - TextMate's HTML Tidy inserted a:

                /*]]>*/

into the <head>.

Daniele

In article 
<1ix8i65.15u7rvv6uxbmlN%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> > <http://dorayme.netweaver.com.au/proc2.html>
> 
> Sadly, display:table-cell means losing control over widths and margins.

When I alter the width of .firstcolumn to smaller, it gets smaller. 
Ditto the other col. When I set width appropriately (eg. .firstcol to 
40%) and alter margins (eg. margin to 20px), the margins take effect. At 
least on FF.

-- 
dorayme

dorayme  wrote:

> > > <http://dorayme.netweaver.com.au/proc2.html>
> > 
> > Sadly, display:table-cell means losing control over widths and margins.
> 
> When I alter the width of .firstcolumn to smaller, it gets smaller. 
> Ditto the other col.

Yes - but only when inside a container with display: table. Otherwise,
width has no effect.

> When I set width appropriately (eg. .firstcol to 40%) and alter margins
> (eg. margin to 20px), the margins take effect. At least on FF

Are you sure? Margins on table cells shouldn't be displayed (and aren't,
in any browser I have). 

Daniele

In article 
<1ixe0f3.xyjw895l1creN%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> dorayme  wrote:
> 
> > > > <http://dorayme.netweaver.com.au/proc2.html>
> > > 
> > > Sadly, display:table-cell means losing control over widths and margins.
> > 
> > When I alter the width of .firstcolumn to smaller, it gets smaller. 
> > Ditto the other col.
> 
> Yes - but only when inside a container with display: table. Otherwise,
> width has no effect.
>

In the above, the width of .firstcolumn is responsive to change. Perhaps 
this example does not suit your purpose, fair enough. I was trying to 
answer your request in your original post.  
 
> > When I set width appropriately (eg. .firstcol to 40%) and alter margins
> > (eg. margin to 20px), the margins take effect. At least on FF
> 
> Are you sure? Margins on table cells shouldn't be displayed (and aren't,
> in any browser I have). 
>
On my Mac FF (and I would sort of bet, all modern browsers - not IE) it 
does what I say above, I did not just say it, I tested it and am seeing 
it right now again. Did you test exactly what I said above in relation 
to the URL above? And are you getting a different result in your FF and 
Opera and Safari?

-- 
dorayme

dorayme  wrote:

> > > When I alter the width of .firstcolumn to smaller, it gets smaller.
> > > Ditto the other col.
> > 
> > Yes - but only when inside a container with display: table. Otherwise,
> > width has no effect.
> 
> In the above, the width of .firstcolumn is responsive to change. Perhaps
> this example does not suit your purpose, fair enough. I was trying to
> answer your request in your original post.  
>  
> > > When I set width appropriately (eg. .firstcol to 40%) and alter margins
> > > (eg. margin to 20px), the margins take effect. At least on FF
> > 
> > Are you sure? Margins on table cells shouldn't be displayed (and aren't,
> > in any browser I have). 
> >
> On my Mac FF (and I would sort of bet, all modern browsers - not IE) it
> does what I say above, I did not just say it, I tested it and am seeing
> it right now again. Did you test exactly what I said above in relation
> to the URL above? And are you getting a different result in your FF and
> Opera and Safari?

I beg your pardon, I didn't notice that you were referring to .firstcol
rather than the table-cell items (which are the ones I'm trying to
control, and which, because they are table-cells, are uncontrollable in
those respects).

Daniele

In article 
<1ixe5k0.1viwjdx1kwe8n6N%real-not-anti-spam-address@apple-juice.co.uk>,
 real-not-anti-spam-address@apple-juice.co.uk (D.M. Procida) wrote:

> dorayme  wrote:
> 
> > > > When I alter the width of .firstcolumn to smaller, it gets smaller.
> > > > Ditto the other col.
> > > 
> > > Yes - but only when inside a container with display: table. Otherwise,
> > > width has no effect.
> > 
> > In the above, the width of .firstcolumn is responsive to change. Perhaps
> > this example does not suit your purpose, fair enough. I was trying to
> > answer your request in your original post.  
> >  
> > > > When I set width appropriately (eg. .firstcol to 40%) and alter margins
> > > > (eg. margin to 20px), the margins take effect. At least on FF
> > > 
> > > Are you sure? Margins on table cells shouldn't be displayed (and aren't,
> > > in any browser I have). 
> > >
> > On my Mac FF (and I would sort of bet, all modern browsers - not IE) it
> > does what I say above, I did not just say it, I tested it and am seeing
> > it right now again. Did you test exactly what I said above in relation
> > to the URL above? And are you getting a different result in your FF and
> > Opera and Safari?
> 
> I beg your pardon, I didn't notice that you were referring to .firstcol
> rather than the table-cell items (which are the ones I'm trying to
> control, and which, because they are table-cells, are uncontrollable in
> those respects).

If I could get a sense of what you really want to do in any realistic 
situation, I would be happy to look.

-- 
dorayme