home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!agate!rsoft!mindlink!a2
- From: Frank@mindlink.bc.ca (Frank I. Reiter)
- Newsgroups: comp.unix.sysv386
- Subject: Impressions of ODT 2.0
- Message-ID: <14604@mindlink.bc.ca>
- Date: 26 Aug 92 18:45:39 GMT
- Organization: MIND LINK! - British Columbia, Canada
- Distribution: world
- Lines: 64
-
- Last week we moved our development efforts from ODT 1.1 to ODT 2.0. So far my
- impressions of ODT 2.0 can be summarized thusly:
-
- ODT 2.1 should be really nice.
-
- I think 2.0 was rushed - I've found it not to be up to SCOs usual standards.
- In less than a week we have found various problems, and SCO has not been able
- to resolve all of them to our satisfaction.
-
- Examples:
-
- - In the installation notes you are warned "It is recommended that the
- development system be installed in single user mode" or words to that effect.
- What it should say is "You absolutely must install the development system in
- single user mode or you may not be asked for a serial number and activation
- key, and the compiler will not run."
-
- I'm told this will be documented this way in the next release.
-
- - We increased NREGION from 300 to 450 to get rid of a region table overflow.
- The kernel would not relink - it complained about "missing required driver
- aud". A call to SCO got us the help we needed - running fixperm changed the
- group on various link kit files and all was okay - but how did it get this way
- in the first place? There are no 3rd party drivers on this machine.
-
- - The compiler gives is spurious warnings on correct C code which calls
- functions by using a pointer to the function. SCO's solution? Use cc -W0 or
- don't prototype the function.
-
- Thanks a lot. We have an application with 150,000+ lines of source code here.
- We rely on prototypes and compiler warnings to aid us in our development.
-
- I'm told this will be corrected in the next release.
-
- - The compiler dumps core compiling modules of that same 150K line application
- if symbolic debugging information is included. SCO's solution? Create Xenix
- format binaries (which requires that we convert all of our libraries to that
- format as well) or don't use symbolic debugging information. This will be
- fixed in the next release.
-
- - We're having trouble using codeview even with the applications that WILL
- compile with symbolic debugging info - the <ENTER> key is not recognized (nor
- will ^M or ^J yield the desired results.) We have a support call from SCO
- scheduled on this one. I have this terrible fear that the answer will be
- "Don't use codeview - this will be fixed in the next release."
-
-
-
- I've been hearing mention of this next release quite a lot this past week.
- What I haven't heard is whether to expect it next week, next month, or next
- year, or how much I'm going to have to pay for it. If others are having as
- much trouble as we are then perhaps ODT 2.0 ought not to have been released
- yet. It has fixed a bunch of annoying problems with 1.1, but the problems we
- have now will have a much greater impact on our productivity. We may go back.
-
- This is only the experience of one site, but so far this site's recommendation
- is to wait for "The next release (tm)". I'd like to hear from other 2.0 users
- who agree or disagree.
-
- Frank.
- --
- ___________________________________________________________________
- Frank I. Reiter Internet: Frank@mindlink.bc.ca
- MIND LINK! Communications Corp. Surrey, British Columbia, Canada
-