On Sun, Nov 14, 2004 at 02:46:47PM -0500, Richard Hayes wrote: > I am investigating a sponsorship project that may get funding. > > What would be the most valuable free /OSS software that does not exist yet?
That's going to be some very subjective answers you're going to get. My suggestion? A quality plugin for Outlook that stores MAPI objects in messages stored on an IMAP server. Like Bynari's Insight Connector, but one that actually works. That would allow any IMAP server act as a replacement for the vast majority of Exchange. It's also a relatively small project, so it has a chance of getting done(ish). > 1. Free Wrap system (superannuation management) > > Wrap accounts let users invest in a wide range of investments through a > single > interface such as the web. It manages all the statuary requirements for > super rules. > > Wrap accounts charge between 0.6-1.0 % of funds under management per year. > With Australian Super Funds having over $600 Billion up management. > Currently about 20% of clients use wrap accounts so the fees are worth up > to $ 1 Billion per year. (Half that would be more reasonable) So what will the software you write actually do? Manage the wrap accounts? Tha value there isn't in the software, it's in the access to the trading accounts and first-level equities traders, and also in the research. You're going to have next to no hope of convincing any of the wrap account managers to take it up, either, because of the legal ramifications of the software screwing up and not doing things right. On the other hand, if you get it written, let me know and I'll pitch it to one of my clients. > Using OpenAdaptor (opensource EAI tool for finance) + struts.apache I do > not > think it would be too hard. ;) Wow, the OpenAdaptor website is buzzword-compliant. I still have no idea what it's actually good for. > eBXML should makes it easier to intergrate legacy systems. Ha ha ha. You're a crack-up, Richard. I'm still trying to explain why sending daily equities price and trading data dumps via broken CSV is a bad idea -- XML still lies deep in the "what the hell is that?" category in a large portion of the financial services industry. > 2. Linux-based Voice recognition software I'm quite surprised there isn't a fair bit of this already out there, with the geekly desire to play with new technology. Might be a nice project, but I suspect it may be more of a research project than product development, unless you hire away a couple of major brains at, say, Dragon or somewhere... > Please comment and if anyone wants suggest a way forward I would be grateful. Bounty for features you want in existing apps. The Gnome experiment along those lines mustn't have totally sucked, because I've seen several more implementations of the same idea more recently. - Matt
signature.asc
Description: Digital signature
-- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
