[462] in Tooltime
Re: Some Scopus questions
daemon@ATHENA.MIT.EDU (Steven Wade Neiterman)
Mon Nov 3 14:20:22 1997
In-Reply-To: <v0302090bb083bd4f9d00@[18.81.0.146]>
Date: Mon, 3 Nov 1997 14:24:13 -0500
To: Barbara Goguen <goguen@MIT.EDU>
From: Steven Wade Neiterman <wade@MIT.EDU>
Cc: tooltime@MIT.EDU
At 12:50 PM -0500 11/3/97, Barbara Goguen wrote:
>- Is there a limit to the amount of text that can be entered into the
>Summary Field?
Speaking from the Oracle angle, the field is 255 characters long. Maybe
David or Peter can say if the GUI has a different restriction.
>
>- Printing is REALLY ugly. When you click on the print button you get this
>printout that's totally unformatted listing all the fields and their values
>from the log you are viewing. I assume that this is because getting a
>screen shot or a more formatted printout was determined to be too
>difficult. I am also assuming that this will be a hot issue when I do the
>training on Tuesday evening so would appreciate any defense amunition
>anyone can offer.
See Rob's mail.
>
>- Can you sort on more than one column with a results set window. It
>appears to me that you can only sort on one, but multiple sorts might be
>nice - first by category then by status for example.
Hmm. When using the GUI, I would think you could only sort on one column
as you do in Eudora or the MacOS Finder. My guess is that sorting on more
than one column would require a script.
>
>- Running Scopus on my SparcClassic is REALLY slow. Running the Find Open
>script took a really long time. Then sorting the result set was reaaly
>slow. It appears to be trying to constantly repaint the screen whenever
>there is a result set window open. Has anyone else encountered this?
Might be interesting if someone could benchmark the NT version with
Solaris. It's possible that the Solaris client is doing something strange,
but a comparison of platforms would give us more data.
>
>- In testing how record locking works, I encountered some strange behavior.
>I was first logged into two different machines, but under my account on
>both. I called up the same log on both. Modified the log different on
>each machine, and was able to save without getting a record lock message.
>When I went to take a look at what had happened, Scopus had actually split
>the log into two with duplicate information except the updates - one record
>had the update from one session and the other record had the update from
>the other session. I then tested record locking logging in as two different
>people (myself and Peter Lee) and it did the right thing - locked the
>second access to the log out.
The database only has two fields to determine if a record is locked. One
field is a yes/no flag, the other field has the username of the person
logged in. The code probably does a comparison of username and knows
nothing if your logged in more than once. This explains the above senario.
This came up in one of the team meetings and the thought was that a person
would not be logged in more than once on different machines.
>
>- What did we ever decide about the use of the child windows and procedured
>for mainframe, network and housecall? Did we decide anything?
Not really sure what your asking. If we are not using mainframe and
housecall, maybe it's best to remove it.
..Steve
