W3C home > Mailing lists > Public > www-style@w3.org > March 2001

Re: css3 :nth-child() WD

From: Robin Berjon <robin@knowscape.com>
Date: Fri, 02 Mar 2001 20:09:33 +0100
Message-Id: <4.1.20010302195217.02950e40@mail.knowscape.net>
To: glazman@netscape.com (Daniel Glazman)
Cc: www-style@w3.org
At 09:25 02/03/2001 +0100, Daniel Glazman wrote:
>Robin Berjon wrote:
>> I totally agree ! cron-style notation is by far easier to read. I also
>> think it's generally far more intuitive than the an+b notation. 
>
>More people in the unix community, probably. Seriously, try to put a
>web author in front of a crontab entry and take a look at his face...
>This is CSS, not a subdirectory of /var/spool

hehe. Seriously, by cron style I didn't mean full-blown cron syntax.
Putting a web author in front of nth-child(2,5,8-11) and nth-child(1,2-*/3)
didn't bring up a strange face. It brought up a "Cool !" and he could
figure out what it meant immediately. Otoh, nth-child(-5n+6) didn't seem to
be as intuitive. Linear sequences aren't hard to understand (at least in
France everybody has been through them, I don't know about other
educational systems), but for many people it's far behind.

>1. I think that Tantek's father's suggestion which is in the current
>   Last Call WD is elegant and simple, for both implementors and users,

I'm not saying that it's inelegant, in fact I like it. I'm just suggesting
that there may be schemes easier to understand for the end user. As for
implementors I wouldn't think it makes much of a difference so long as it's
easy enough to massage. Both linear and cron are ok in that respect. I've
been thinking about adding a position_matches($position) to SAC's
PositionalCondition that given a node's position would return a boolean
indicating whether the condition is true or not. Shouldn't be too hard in
any of those cases.

>2. I don't want to see a coding whitespace in the argument of a pseudo
>   unless this whitespace is a descendant combinator,

My gut feeling agrees with this point, but I'm not sure why. Do you have
any specific reason ? Anyway, this doesn't concern so-called cron notation
which is what I have in mind.

>The an+b notation is in between cron's notation and Sicking's proposal. I
>am not at all in favor of a change.

I'm not saying that I think the current linear way should change. I'm not
especially opposed to it. Just my E0.02 about how I feel cron style
notation is simpler, and not just to seasoned Unix sysadmins.

-- robin b.
Prediction is very difficult, especially of the future. -- Niels Bohr
Received on Friday, 2 March 2001 14:07:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:08 GMT