It looks like your responses, including mine here, are unanimous - it's preferred to store blobs outside the DB and metadata inside. I see no benefits to storing blobs in the DB and lots of reasons not to do so. I wrote a blog a while back on signature capture, which I implemented much like Rob Porter did his document scanner. In addition to capturing images, a "real" signature scanner collects biometric data. It's all the same, and best kept outside the database, indexed and referenced by the database. nospamNebula-RnD.com/blog/tech/2009/09/signatures01.html
As far as document management in general, my wife works in the legal department for a large USA company. They're looking for a new system to store contract source/Word documents, to scan and generate PDFs, OCR and index TIFF/PDF, and to link all of these with clients, trading partners, employees, events, legislation requiring specific language, other documents with with similar language, and other databases. They need workflow, historical reporting, and of course high security. I know all of that can be done very well in a Pick environment - if we just separate frilly document-related functions from data-related functions. Software like this easily sells for $100K upwards, but if you look at those packages, they do a lot of the frilly document processing (which is trivial and very low-cost for anyone who knows how) and minimal data processing. Post-sale development work to make these systems do what's required can easily double the package cost. Writing a new system like this in U2 would take many months but could yield a really nice package, more malleable, and at a much lower TCO. What we generally see (as in this thread) is that most Pick people are strong on the DB but not so strong on external interfaces, which is what people see and pay the big bucks for. If anyone is familiar with a Pick-based package that fits the above description, please contact me. If you have the budget to create one in preference to buying one, I'm open to that as well. Thanks! Tony Gravagno Nebula Research and Development TG@ remove.pleaseNebula-RnD.com remove.pleaseNebula-RnD.com/blog Visit PickWiki.com! Contribute! http://Twitter.com/TonyGravagno > Does it make any sense to use a U2 database to save scanned > pdf's? That is... > > Does it make any sense to save small blobs in a U2 data base? > > Would a Type 1 file be best? > > Are there any advantages to using MS-SQL, MySQL, or some > other database technology? _______________________________________________ U2-Users mailing list [email protected] http://listserver.u2ug.org/mailman/listinfo/u2-users
