Jon Edwards wrote: > > Thanks Tim for a very interesting post! > > Re. the Palm/PDA problem,
I should have been a bit more specific re PalmPilots. The problem is not that technological solutions don't exist, its just that we couldn't find one which didn't involve either considerable expense or considerable effort in learning and supporting a new, proprietary development environment to provide a means by which people could browse and edit current issues/event on their PalmPilots while off-line, and then synchronise with the main database when they next dock the PalmPilot. The same issue applies to laptops taken on the road, but there you can use Python and the same database engine as Roundup, so the problem is just one of synchronising the database on teh laptop with the main Roundup database - not an insurmountable problem - and also providing a non-Web interface to Roundup since laptops on the road are not yet Web-connected (at least ours aren't) - we are thinking about wxPython for this GUI, which of course could also be used on desktop clients as well. > I've never used one, but can't you adapt the > web/Zope front-end of Roundup for PDAs? I found a couple of articles on > getting Zope to output a WAP-compatible version of your pages, for example - > > http://www.squishdot.org/948313901/index_html > http://www.zope.org/Members/Duncan/wap_howto > > - perhaps you could do something similar? WAP impiles a) the PalmPilots are WAP enabled with a wireless connection - they aren't b) if they were, we would need a WAP channel via a mobile phone network operator - such channels are not cheap. > > Also stumbled across "Python for the PalmOS" > http://www.isr.uci.edu/projects/sensos/python/ > > or Pippy - http://www.endeavors.com/pippy/ Pippy works as a "programming doodle pad" but is not suitable for real-life use, not because of bugs, but due to intrinsic limitations imposed by the PalmOS environment: no support as yet for a forms interface, no support for floating point maths (and hence no date/time support) etc. In fact, on closer examination, the PalmPilots are not very desirable application development platforms. They have slow (16 MHz) CPUs without any hardware floating-point support (must be emulated in sofwtare - remember the 8087 chip which you needed to add to your PC in order to make large spreadsheets recalculate more quickly...there;s no such thing for a PalmPilot), the operating system has no concept of a file, and all the (usable) development tools are PalmOS-specific. Yuk! But I still love my PalmPilot, at least until Santa brings me a Linux-based PDA such as a Yopy or a Sharp Zaurus or a Compaq IPaq (with WindowsCE replaced by Linux). > > and "Palm-Linux integration with Pyrite" - > http://www-106.ibm.com/developerworks/linux/library/l-pyr.html Yup, Mertz's conclusion is that Pyrite is not ready for production use, and it doesn't seem to have progressed since he wrote the article. There are probably hundreds of thousands of open source projects in existence, but only a tiny fraction of them are of any use to anyone other than the original author(s). Sometimes the sheer number of open source solutions, or half-finished solutions to a particular problem can make the task of evaluation (of those solutions) very time-consuming. It can take half a day to decide whether an open source project is a serious and well thought out bit of work or just someone fooling around for fun and/or self-education (to which I have no objection, as long as the results are clearly billed as such). Ultimately, some sort of organised peer-review system for open source software in various categories (such as health/medical) would be useful in order to distribute this task of evaluation. Tim C
