This embedded SVG seems to be tough for my laptop's browsers to render quickly. Is it something I said?
There is poorly applied styling concerning this and other SVG entries. I also adjusted the line length so that is why it renders (on my display, at least) too big for the other items as they are laid out.
Also, the 8-bit Braille encoding is temporarily inaccurate.
Right-click and zoom out if the Braille is too big for the page, and hopefully you won't see a nasty missing bit of content that should have just come in from the outside of the visible part of the designated SVG viewport.
It is perhaps better to render several inline SVG objects in one entry─maybe word-wrapping can cease to be a problem that way...
I hope Opera 9 is better than 8.51 at this sort of thing.
Here's a preliminary effort at Braille in SVG. The file was generated using a Python script. The next step is line breaks at every nth character or the first preceding white-space character. For more appropriate accessibility, a format to be parsed and rendered by a mechanical or other Braille reader will be necessary. Also, an animated format displaying individual Braille words in sequence may be useful too. Though I don't have any idea right now about how or if animated tactile displays can handle this kind of thing, a product with such capabilities would be rather neat.
For now (to see everything) right click the image and zoom out. The full text is: "this is a first braille test svg generated by python."
When the Python script improves it will be made available.
To the maximum extent possible, Scale Vector Graphics (SVG) should be accessible to everyone. For SVG to be accessible to non-capable browsers ─ and more importantly, useful to anyone using its alternate text ─ seriously meaningful and flexibly produced prose needs to be parsed into the altenate text node or attribute.
This can be done with Python. I'll post more when I have looked into this issue a bit further.
The SVG image file below displays correctly in Opera (OS:Kubuntu Linux) without plugins but incorrectly (if at all) in Mozilla, MSIE, FireFox and Konqueror 3.5 with or without plugins, according to my inexpensive (read: non-extensive) tests.
This demo is intentded to apply priniples from Tufte's useful publications on informative displays. Animation might be a relatively good way to extend this particular example. There are some comments to this end in the SVG source. It is almost marked up correctly now...
The two graphic representations of equity price fluctuations should contain enough information, but as little as necessary to allow the representations to be quickly informative, conveying the specific information that one share should be held but the other can be sold. This is done using colours (green to sell, red to keep), labels, numbers, the trigger range of values over time (shading) and the line plot of the price over time.
It might be interesting to have options available to display scale vector graphics of accumulated daily uptime (that is, time logged into my.opera) on user pages. By default, this would be as a background layer beneath the days of the displayed month, but also available to be styled by user CSS.
To the extent that this should generally be considered private data, users would be required to opt-in to such a service.
The options might include different formats, including but not limited to: ·single vertical bars per day, whose height indicates total time logged in over the 24-hour period. ·shaded horizontal bars indicating actual times logged in ·shaded horizontal bars indicating expected active periods, ie. lunch-breaks, time away from a web interface
Cumulative time might be preferable as it would take up less of the users' diskspace. Control over the amout and type of log-in data made available to the served pages ought also to be available to the user.
Another options could be the option of when days begin according to the user, such as +1 hour GMT, -15minutes EST, or +16hours-11minutes+12seconds. I favour overkill on arbitrary functionality such as minutes and seconds, because then the user─whether they choose to use the functionality or not─can choose their level of control and involvement.
Reasons this can be considered a productivity aid:
·Users can quickly see in context visual information about how much time they spend logged in, and can compare it to what they perceive they have got done and when.
·Such graphs aid pattern recognition and can thus support behaviour enhancement or planning and control.
·Information enables, whether it is consciously observed or passively absorbed.
Implementation would be a simple parser script filtering data from user logs whenever a change of state occurs.
Edward Tufte has one main idea, and it is simple, but powerful: show people as much data as possible with as little ornamentation as possible. Let the data speak for itself.