I didn't mean to start the "editor" discussion, but it was informative.
Anyway, AFAIK UniData supports only Prime-type dictionaries, while UniVerse supports both Pick and Prime style dictionaries. If converting from a Pick system, UniVerse is easier from a dictionary standpoint. Dick Kryka Director of Applications CCCS of Greater Denver, Inc. Paragon Financial Services 303-632-2226 [EMAIL PROTECTED] -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of David Beahm Sent: Friday, September 24, 2004 8:51 AM To: [EMAIL PROTECTED] Subject: Re: [U2] Comparison Unidata and Universe I didn't voice my opinion earlier because I know this list has a lot of developers with more experience in both than I do, but here goes... <flameproof suit on> One of the most critical elements in the MV market has been cross-system compatibility. Companies have invested in systems and like what they've got -- they just want to move to a newer box and/or add new capabilities. UniData and UniVerse, having been originally developed by different companies, meet this need in different ways. In UniVerse, one creates an account and decides what "flavor" the account will be. Largely this is a matter of command line and BASIC syntax, although it also affects details such as how select lists are handled in certain situations, etc. There are some ways to add elements from alternate flavors, but things are generally set once you create the account. In UniData, the different syntaxes and behaviors are intermingled but can be specifically accessed by setting runtime variables (UDT options, BASICTYPE, ECLTYPE, using lowercase commands) at the command line or within programs. If you wanted to bring together code from a variety of systems, UD would probably support that better than UV. Even so, the syntax is about 80% identical between them. Each has unique methods for file processing, which comes dramatically into play when file sizes exceed 2GB. AFAIK UniData has one method for dealing with large files (make them dynamic), while UniVerse has a few (part files that break one file into many, 64-bit file addressing, maybe something else?) The UniData debugger is significantly better than UV's, however I believe if you use UniDebugger this point becomes less significant. Both provide nearly identical support for UniObjects in developing VB/VCC/.Net/Java applications. UD has a GUI tool for defining ODBC schema while UV relies on DICT entries and a few other settings. It seems to me that UV tries to figure things out for ODBC on the fly, while UD decides ahead of time and applies less forgiving but faster logic. There are dozens of little things that one offers that the other doesn't, things one person loves and another hates. For instance I wish UD supported Q-pointers (shortcuts to files in other accounts), and I wish UV allowed sending LIST output straight to a file the way UD does. In both cases I made workarounds for these, but it's always nicer to have these things out of the box. Fortunately, features added to one are also being added to the other, so IBM is doing a nice job with that. <flameproof suit off> I hope that helps clear things up, and I hope I didn't get too much wrong. ;^) I expect others will be along shortly with corrections/explanations/rationalizations/pontifications. Best, David Beahm Bob Little wrote: > [EMAIL PROTECTED] wrote: > >> Since this thread seems to be going on forever.. >> >> > > Yes, it does. But we've strayed considerably from my original question > which was: > > What's the difference between UniVerse and UniData? ------- u2-users mailing list [EMAIL PROTECTED] To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [EMAIL PROTECTED] To unsubscribe please visit http://listserver.u2ug.org/
