We have been putting slides and sometimes audio of seminars onto the
web for a few years now. The script that I wrote to produce the webpages
is starting to creak as more demands are placed on what it must do.
At some point I shall have to find time to rewrite it cleanly.

Having just discovered this newsgroup, I should be interested in
critique of the output in its current state, particularly usability
and accessibility.

A reasonably representative seminar is:
http://www.newton.cam.ac.uk/websemi...rt/0803/forbes/

The top-level is:
http://www.newton.cam.ac.uk/webseminars/

Some shortcomings I already know about are:

* No height or width specified for images.
I didn't know how to extract this data from files when I
wrote the script. It should be a relatively simple addition.

* Most images have been scaled to fit into an 800x800 bounding box.
This, however, may entail scrolling on a 1024x768 display.
Perhaps a 720x720 bounding box would be better?

* Usually, full-size images in the html pages are linked to themselves
(to enable one to remove the surrounding clutter). On rare occasions
images are linked to much larger versions. However, no indication of
this is given. I'm not even sure how one would do so.
A comment about size in the ALT tag?

* The "help" page is getting increasingly out of date.

* No search form.
At the moment, the only way to find a talk is to know where to look.
It's on my list of things to do.


I am also interested to see examples of how other places have gone about
making presentations available online. About the only site I know of is:
http://online.itp.ucsb.edu/online/
Pointers to other sites would be appreciated.


Once we have a convenient way to record timing details, I hope to be able
to provide a SMIL wrapper for synchronised streaming of audio and images.
Should I assume critique of SMIL is outside the scope of this group?


thanks,

-jonathan

--
Jonathan H N Chin, 1 dan | deputy computer | Newton Institute, Cambridge, UK
<jc254@newton.cam.ac.uk> | systems mangler | tel/fax: +44 1223 767091/330508

"respondeo etsi mutabor" --Rosenstock-Huessy

Jonathan H N Chin wrote:

> We have been putting slides and sometimes audio of seminars onto the
> web for a few years now. The script that I wrote to produce the webpages
> is starting to creak as more demands are placed on what it must do.
> At some point I shall have to find time to rewrite it cleanly.
>
> Having just discovered this newsgroup, I should be interested in
> critique of the output in its current state, particularly usability
> and accessibility.
>
> A reasonably representative seminar is:
>     http://www.newton.cam.ac.uk/websemi...rt/0803/forbes/
>
> The top-level is:
>     http://www.newton.cam.ac.uk/webseminars/
>
> Some shortcomings I already know about are:
>
> * No height or width specified for images.
>   I didn't know how to extract this data from files when I
>   wrote the script. It should be a relatively simple addition.
>
> * Most images have been scaled to fit into an 800x800 bounding box.
>   This, however, may entail scrolling on a 1024x768 display.
>   Perhaps a 720x720 bounding box would be better?
>
> * Usually, full-size images in the html pages are linked to themselves
>   (to enable one to remove the surrounding clutter). On rare occasions
>   images are linked to much larger versions. However, no indication of
>   this is given. I'm not even sure how one would do so.
>   A comment about size in the ALT tag?
>
> * The "help" page is getting increasingly out of date.
>
> * No search form.
>   At the moment, the only way to find a talk is to know where to look.
>   It's on my list of things to do.
>
>
> I am also interested to see examples of how other places have gone about
> making presentations available online. About the only site I know of is:
>     http://online.itp.ucsb.edu/online/
> Pointers to other sites would be appreciated.
>
>
> Once we have a convenient way to record timing details, I hope to be able
> to provide a SMIL wrapper for synchronised streaming of audio and images.
> Should I assume critique of SMIL is outside the scope of this group?
>
>
> thanks,
>
> -jonathan
>
It's behind the times.  It'll be easier to
send you off for a read than to pick away at
details when a paradigm shift is really needed.

Getting away from deprecated markup,
and into semantic markup:
http://www.alistapart.com/
http://www.realworldstyle.com/
http://w3c.org
http://www.webstandards.org

Frames on the internet:
http://www.html-faq.com/htmlframes/?framesareevil
http://www.useit.com/alertbox/9612.html
http://www.highrankings.com/framedsite.htm
You're generating the pages,
so there's really no reason to be
using frames for this kind of site.
--
mbstevens
http://www.mbstevens.com/preprocessor/
http://www.mbstevens.com/howtothumb/

On 14 Sep 2004 18:38:20 GMT, jc254@newton.cam.ac.uk (Jonathan H N Chin)
wrote:

>Having just discovered this newsgroup, I should be interested in
>critique of the output in its current state, particularly usability
>and accessibility.
>
>A reasonably representative seminar is:
>    http://www.newton.cam.ac.uk/websemi...rt/0803/forbes/

I've seen much worse. You've refrained from the current silly fad of
tiny low-contrast text: please don't change that!

A few thoughts:

- ask yourself if you really need the frames (though I think you may
have better reasons than some sites for keeping them).

- you have provided an option to remove frames; I can't immediately see
how to get them back if one wants the navigation frame again.

- you use <H2> to make the name of the author larger, but it isn't
really a heading. Better to use CSS to change the text size.

- in e.g.
http://www.newton.cam.ac.uk/websemi...forbes/008.html
you have a lot of text rendered as an image. You could make this into
actual text in a table, and then it would (a) be more accessible
(b) load faster and (c) everyone could scale the text to the size they
like to read. Similarly in section 006 all the text could actually be
real HTML text.

- you use &nbsp;&nbsp;&nbsp; etc for indentation; if you use CSS
margin/padding the result is simpler and easier to maintain; similarly
for use of <BR><BR> for vertical spacing.

- given the small size of the pages here, I don't think I'd bother
having the menu at both the top and bottom of the page - but that's a
personal preference.

I haven't tried the multimedia aspects.


>* Most images have been scaled to fit into an 800x800 bounding box.
>  This, however, may entail scrolling on a 1024x768 display.
>  Perhaps a 720x720 bounding box would be better?

Or even a bit smaller if the images aren't too detailed. I have a
1024x768 screen, but my typical browser window is about 750x600. In
general, make each image big enough to clearly convey the information on
a typical monitor; don't worry about all images being the same size.

>* Usually, full-size images in the html pages are linked to themselves
>  (to enable one to remove the surrounding clutter).

I must say I found that a bit confusing when I clicked on the page to
get back to it (from another window that was obscuring part of it), and
the page suddenly changed. I suggest you don't do that. If you feel that
the page content isn't sufficiently set off anyway, separate it by means
of an extra horizontal rule, or by giving the menus a different coloured
background (a subtle pastel tint would suffice).

>On rare occasions
>  images are linked to much larger versions. However, no indication of
>  this is given. I'm not even sure how one would do so.
>  A comment about size in the ALT tag?

Better to make it a separate link: "Large version of diagram (700 kB)"


Hope this helps.
--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

Stephen Poley <sbpoleySpicedHamTrap@xs4all.nl> writes:
>On 14 Sep 2004 18:38:20 GMT, jc254@newton.cam.ac.uk (Jonathan H N Chin)
>wrote: 


>- ask yourself if you really need the frames (though I think you may
>have better reasons than some sites for keeping them).

I don't normally like frames myself.
However, it seemed the most practical way to keep the menu of
thumbnails around without having to redraw them all the time.
The thumbnails aren't always very distinctive, but when they are,
I suspect they are more useful than memorising numbers.


>- you have provided an option to remove frames; I can't immediately see
>how to get them back if one wants the navigation frame again.

The link is "use_frames" on the page you reach from "thumbnails",
"thumbnails" being the equivalent of the menu frame.

I don't know how to sensibly jump back to the framed version
of a page as opposed to the introduction, so I hadn't felt it was
particularly confusing to not have a "use_frames" link cluttering
each individual page. It's something to think about though.


>- in e.g.
>http://www.newton.cam.ac.uk/websemi...forbes/008.html
>you have a lot of text rendered as an image. You could make this into
>actual text in a table, and then it would (a) be more accessible
>(b) load faster and (c) everyone could scale the text to the size they
>like to read. Similarly in section 006 all the text could actually be
>real HTML text.

Yes. In an ideal work, the text would be transcribed from the images.
As and when software to reliably do so becomes available, I shall be
happy to incorporate transcriptions.

It's hard enough just getting slides from the speakers in the first place.
If they were required to also transcribe them to text from whatever dodgy
presentation package they used (TeX, powerpoint, pdf, keynote, handwritten
overheads, blackboard, ...) I doubt we'd receive anything!

And our manpower is finite, so we can't do it for them.
However, if enough users demand it, it may be possible to get funding.


>- you use &nbsp;&nbsp;&nbsp; etc for indentation; if you use CS
S
>margin/padding the result is simpler and easier to maintain; similarly
>for use of <BR><BR> for vertical spacing.

I still use Netscape 4.7x as my main browser, with javascript turned off,
so I like pages to work without CSS. Also, I don't know any CSS yet...

Come the rewrite, I'll certainly allow for CSS, but I'd like the pages
to remain reasonably laid out even without it. That said, a lot of the
complexity probably comes from centering the columns. If I left things
left-justified, I could no doubt do away with much of the dodgy markup.


>- given the small size of the pages here, I don't think I'd bother
>having the menu at both the top and bottom of the page - but that's a
>personal preference.

Some talks have more images. For example:
http://www/webseminars/mondays/2000/11/06/boyland/

 

Also, the script was written with picture galleries in mind,
although so far there are only a handful of galleries online.


>I must say I found that a bit confusing when I clicked on the page to
>get back to it (from another window that was obscuring part of it), and
>the page suddenly changed. I suggest you don't do that. [...]

Thinking about it, I would guess most browsers have a "view image"
or "open image" option, so people who want the functionality may well
have it already even without the self-links.

 
>Better to make it a separate link: "Large version of diagram (700 kB)"

Yes. This sounds like a good idea. There is no real need to link
images at all if there isn't a larger version available.
And separate links allow for the option of multiple resolutions.


thanks for the suggestions,

-jonathan

--
Jonathan H N Chin, 1 dan | deputy computer | Newton Institute, Cambridge, UK
<jc254@newton.cam.ac.uk> | systems mangler | tel/fax: +44 1223 767091/330508

"respondeo etsi mutabor" --Rosenstock-Huessy

mbstevens <NOXwebmasterx@xmbstevensx.com> writes:
>Jonathan H N Chin wrote: 


>It's behind the times.  It'll be easier to
>send you off for a read than to pick away at
>details when a paradigm shift is really needed.

As I say, I only know of one other site presenting similar material.
To motivate the subject (as mathematicians say), it would be useful
to have a concrete example of a site that isn't behind the times.
Where are they hiding?

Thank you for the list of links. Your one about thumbnails looks
as though it will be particularly helpful come the rewrite.


-jonathan

--
Jonathan H N Chin, 1 dan | deputy computer | Newton Institute, Cambridge, UK
<jc254@newton.cam.ac.uk> | systems mangler | tel/fax: +44 1223 767091/330508

"respondeo etsi mutabor" --Rosenstock-Huessy

Jonathan H N Chin wrote:

> mbstevens <NOXwebmasterx@xmbstevensx.com> writes: 
>
> 
>
> As I say, I only know of one other site presenting similar material.
> To motivate the subject (as mathematicians say), it would be useful
> to have a concrete example of a site that isn't behind the times.
> Where are they hiding?

>

I suppose the basic problem is how to display
scientific images along with explanitory text -- I think
it's too much to expect to find good webpage coding within
an extremely narrow field.  But let's look at a few sites that
have to work with _generally_ similar problems.
________________
Well you're right, even Wolfram's Mathematica site
http://www.wolfram.com/
...doesn't validate this evening, is heavy on javascript,
and is inflexible about rearranging text to fit browser
size.
________________
Even Britney's guide to Semiconductor physics is not
exactly a paragon.
http://britneyspears.ac/lasers.htm
_______________
But, here:
http://www.scss.com.au/family/andrew/astro/
validates nicely and functions well --
without the need for frames.
Grab it by a corner and pull the size of the
window back-and-forth.  See how everything adjusts to the
window size.
--
mbstevens
http://www.mbstevsn.com/mscope/