[Haskell-cafe] Haskell Weekly News: Issue 187

Rogan Creswick creswick at gmail.com
Thu Jun 23 23:53:01 CEST 2011

On Thu, Jun 23, 2011 at 2:26 PM, Albert Y. C. Lai <trebla at vex.net> wrote:
> I hate the borrowed academic practice of saying [0] and giving the URL two
> hundred lines later. It worked great on paper in hands because I could stick
> my finger to the paper to remember where to return. It also works great on
> real HTML documents because browsers have a "back" button for the same. It
> completely fails in plain text email because I can't stick a finger and I
> can't press the "back" button. Which one is the bigger disruption:

This depends entirely on why you are reading the content -- something
that there is no consensus on :)

> in-situ long URL that makes me skip oh two lines to continue with the main
> text?

This is substantially more disruptive than two lines when reading on a
cell phone.  Lines that are not intended to wrap in the textual layout
end up wrapping with longer URLs, making it more difficult to figure
out where the next line of actual text continues.  Longer URLs also
create larger areas that you can't touch to scroll a message.

Short, obfsucated, urls may direct you places you don't want to go,
but I fail to see how that concern applies to HWN: since each url is
accompanied by a description of its content, that seems to obviate the
need to see the actual url.  In most cases, the text also indicates
the domain that you will visit, so you can avoid supporting
stackoverflow with page impressions if you wish (for example).

Eventually this should just be a client-side rendering preference, but
we aren't quite there yet.

In any case, I think that's a pretty complete description of my
perspective/motivation (not that I'm strongly motivated), so I'm
bowing out to watch :)


> or the URL postponed by two hundred lines so I have to first remember
> it is [1] not [0] this time, then scroll down several pages to hunt for the
> URL, and then... I forget where to return to?
> If people want short URLs, I don't mind that either, but I'm picky on how
> they are shortened. The shortener should offer the option of showing me the
> original URL and waiting for me to go ahead or abort. As far as I know this
> means tinyurl.com only.
> Thank you for bearing with my rant.
> [0] No URL for this.
> [1] No URL for this either.
> _______________________________________________
> Haskell-Cafe mailing list
> Haskell-Cafe at haskell.org
> http://www.haskell.org/mailman/listinfo/haskell-cafe

More information about the Haskell-Cafe mailing list