Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Thomas Mercer-Hursh

Pages: [1] 2 3 ... 26
1
Sounds like a plan...

2
Suggestions and Requests / Re: drawnativeshape() - extension
« on: May 18, 2010, 04:49:16 am »
Got my vote Paulo ... added to the pages of requests for improving shape scripts.

Isn't it time these got some attention?  It just seems that a small burst of development could really alter our capabilities here.

3
Suggestions and Requests / Re: Half baked idea re reallly big classes
« on: December 08, 2009, 11:17:51 am »
People keep telling me that one shouldn't have that sort of really big class ... having one is a sign that one hasn't decomposed the model properly.  So, I suppose one answer ... but certainly not the one you would prefer ... is "don't do that". :)

It does seem to me to be problematic to have screen and print different.

4
Suggestions and Requests / Re: Punch Card
« on: May 28, 2009, 05:01:16 am »
Sounds like a horrid level of micromanagement to me.  I can't imagine getting any meaningful data from it.  Among other things, it encourages people to make assignments that aren't meaningful.

Reminds me a bit of many years ago when I was doing some computer system acquisition consulting for a job shop manufacturer.  They were explaining the issues they had because the system for tracking what job a person was working on during the day was separate from the system for tracking who had worked on the job, i.e., timecard versus job traveler.  One can't lie on a job traveler because the card is following the job so there needs to be a handoff from X to Y to Z as different people work on it.  But, with their system at the time, there was no automatic tie to the person's timecard.  During the discussion they pulled a random card for a person they thought was a problem and then pulled the travelers for the jobs he said he was working on.  The travelers had only about 50% of the time on the timecard!

If anything, I would think the place to control this would be the source code management system,  Check it out, work on it, and check it in.

5
Suggestions and Requests / Re: Punch Card
« on: May 27, 2009, 05:24:36 am »
Are you looking for timekeeping?  Authorization?

6
Suggestions and Requests / Re: Sharing ShapeScripts
« on: May 04, 2009, 08:29:15 am »
For this concept to get very far, we really need to see some of those many enhancements to shape scripts we have asked for.  I see no reason why shape scripts can't be extended ... and not with huge effort ... such that everything we see is basically the result of an underlying shapescript rather than something hard coded.  And, once we have that, then we can have source for them, subclass them, etc. in the way we have been requesting for some time.

7
Suggestions and Requests / Re: SQL UPDATE Capability
« on: May 12, 2009, 04:44:10 am »
As near as I can tell, if you get in there and muck around in SQL, then you have all of the power and danger of complete access to everything.  

What I did was to make a series of queries to notice what changed around a manual change and then replicated that in SQL (or, actually ABL in my case since I was using an OpenEdge database as a repository).  There are some clues in the class model, but careful attention to actual experiments is advised, as is lots of backups and checking.

8
Suggestions and Requests / Re: SQL UPDATE Capability
« on: May 08, 2009, 09:33:06 am »
Built-in tool based update is naturally preferable since it is presumably safe and structured for the likely purpose.  If you are daring, though, there is nothing to keep you from using SQL to update the model ... everything is just a value in a table in the end.  I build models from scratch and do ex-EA analysis on them, so no reason why you can't puzzle out some mass updates ... as long as you don't mind working with limited documentation.  I.e., take lots of backups.

9
Suggestions and Requests / Re: ALT-F4 just quits the application
« on: January 10, 2009, 11:29:28 am »
But, as noted, filing the request doesn't invalidate discussion here.  In addition to workarounds and alternatives, one might find from the discussion that one didn't really want what one thought one wanted or one might realize that the current thought was only one piece of a bigger issues ... e.g, anything dealing with shapescripts.

10
Suggestions and Requests / Re: ALT-F4 just quits the application
« on: January 09, 2009, 06:45:43 am »
Because posting here you are depending on someone at Sparx to read it and copy it over to the enhancement request system, but if you log the request it will flow directly into that system.

11
Suggestions and Requests / Re: Shape scripts for non-stereotyped items
« on: September 03, 2008, 01:51:13 am »
Go for it!  If added to the other shape script things we have asked for this could add a lot of power.

12
Suggestions and Requests / Re: Paste Position
« on: August 24, 2008, 02:15:26 am »
I suspect that the problem here ... common to a great many products, is that there isn't really any mechanism for positioning the cursor in open space.  Yes, I know it is possible to detect the current mouse position, but that would have its own issues ... like what to do when it is not in the panel where the paste should go.

13
Suggestions and Requests / Re: Support Case number
« on: July 10, 2008, 07:25:00 am »
I thought of also mentioning bug tracking and enhancement requests (with voting), but thought I would focus on one small incremental change at a time. :)

14
Suggestions and Requests / Support Case number
« on: July 10, 2008, 05:05:19 am »
I don't know what Sparx is using internally to track the resolution and status of support cases, but I would like it a lot if when one filed a support case, one got back some kind of identifying number.  There are times that I find that a case goes back and forth a bit and then there is a lull ... sometimes it is just a gap and then picks up again, perhaps after someone has had the time to do some testing .... but other times I am left wondering whether the case has been marked resolved and I didn't know about it.  It would be very handy to have a clear reference number so that one could ask about status and feel reasonably sure that one was talking about the same case, especially if one happened to have more than one issue open at the same time.

I would also appreciate it if, when a case is closed, that a clear simple message to that effect is sent with the status.  E.g., "it's a bug and will be fixed in a future release, so there is nothing we can do about it now" or "We can't reproduce the problem so we don't know what to fix" or "Problem seems to have gone away for reasons which are not clear" or "Solution supplied and effective".  Then, I would know whether or not I agreed!

15
There is something ironic in this having become a Hot Topic thread ...

Pages: [1] 2 3 ... 26