[sword-devel] New SWORD website - update

Peter von Kaehne refdoc at gmx.net
Sun Dec 28 18:48:41 MST 2008

re module stats

I have been searching widely tonight again through the website with
regard to the module stats but remain without answers.

How are the module stats produced? At the moment it appears that the
script is not working - the stats have not changed since ages, the file
is untouched since 15/12, but that was the server move. I am now quite
clear in my recollection that the KJV stat reads as 8624 downloads since
at least some time in late November.

I really need them in a more accessible format. Dropping a fully fixed
html table as currently done is not so useful. A single text file with
tab separated values would be a lot better.


re 1 line vs 2 line menu bar.

I have been given some grief regarding my choice of a single menu bar.

I think I have an acceptable solution - the real problem is to have
narrow screens and shallow screens (wide screens or even worse netbooks)
equally well accommodated. Some compromise is probably inevitable

The previous solution with 2 bars in a table design creates a lot of
redundant space which is a real nuisance and space hog on a shallow
screen, while working pretty well on a narrow screen. I am working a lot
on acer aspire and this annoys badly on there. It is also totally non

My last solution of a single table row works well on shallow and normal
screens, but causes scrolling on very narrow screens.

I think the best solution is to move to an inline list which is styled
as similar as possible to the current (or old) table bars. This will
look as a single long menu bar on a normal or shallow screen. but will
stack up in two or more lines on very narrow screens or with massively
enlarged fonts. The only problem I encounter is that list elements (i.e.
menu items) can get broken up during the wrapping process. The xhtml
nowrap="nowrap" attribute appears to have little effect on Firefox.


re news on new sword front page.

This is now in the process of getting implemented. the problem with that
is obviously that the site will look rapidly dead if no news is getting
added. There are advantages to going completely static - it lowers the
maintenance drastically. I do like news on main pages, but i really
dislike old news prominently displayed. And we are (were?) really bad on
that. I guess the people who want the news up front will have to take
care that it stays up to date.


re styling and coloring.

This remains a contentious issue clearly. I do not expect anyone to like
my design choices - though some seem to do so.

Basically I am moving as much as possible out of hardcoded "styling" -
added white space, forced formatting, table designs - etc into CSS
styles.  In the process I implement CSS styles as I think is
appropriate, but hopefully someone else can eventually create some
alternative styling which has more widespread resonance. The point of my
rewriting is not so much to make it pretty but to make it widely usable
(see below two examples) and make it easy to make better. A one page A4
more or less universal style sheet is a lot easier to fix and improve
than the previous hardcoded way.

The current/old design is breaking badly on narrow devices while being
annoying on shallow devices (see above) - the module library  page
requires sideway scrolling from 800 px downwards, while mine currently
will display well on ~400 px or probably  less - particularly once I fix
the menu all over in the way I suggest above. At the same time my pages
are largely not more than a single screen on a normal screen.

The other aspect I tried to take care of is color blindness - and I do
not mean the inability to choose nice colours, under which many of us
men including probably me suffer - but red-green blindness which makes a
lot of designs hard to follow. A huge proportion of all men have some
degree of red-green weakness, some minor, some profound.

One of the list members with significant weakness has advised me
privately of my color choices until the links etc work for him
(invariably it is a "him"). Wrong choices make links invisible and text
unreadable. Interestingly - and I speak as a doctor here - most men are
simply unaware of their own weakness in this area - you need fairly
profound levels to actually notice, but even weaker levels cause grief
on websites. So, you might not like my choices, but they seem to work
for the brother concerned (and the millions like him) and you need to do
better on this count before you criticise.

In Him


More information about the sword-devel mailing list