|
JCapper Message Board General Discussion
|
| Home |
Register
Log In |
| By | Search | |
| TPW 12/26/2008 9:58:09 AM | Jeff, How difficult would it be to add a "search" feature to the board? We've accumulated enough posts that it makes it difficult to find information from prior months. Terry
| |
| jeff 12/26/2008 11:38:58 AM | Search IS something I want to implement. But it's not likely to happen within the next 60 days. Devoting time to coding out a search engine for the message board takes me away from working new ideas into the JCapper program. Without spilling any beans I have to humbly say there are some kick ass additions on the way. And that's where my focus is until they're ready. Tha's all I'm sayin' for now. < G > -jp .
| |
| TPW 12/26/2008 10:59:47 PM | I understand. You need to slow down on the kicka** ideas, as I can barely keep up with the recent additions. Terry
| |
| ryesteve 12/27/2008 9:04:59 AM | Re: the new additions... are the loose ends from the last update higher or lower on the to-do list?
| |
| jeff 12/28/2008 12:08:25 PM | Steve, I should have a useable daily playlist file in the next program update. -jp .
| |
| ryesteve 12/28/2008 8:06:52 PM | Cool! What about the AFR rank and post time issues? Did you see the same things I was seeing?
| |
| mikejlb 12/28/2008 9:04:42 PM | I used to convert the old Jcapper forum to PDF so I could search key words. This website only converts the 1st page to PDF. If there's a way to convert this site to PDF, I'll convert it for guys that want to search key words.
| |
| jeff 12/28/2008 11:17:51 PM | Mike, The ability on the old (I n v i s i o n) site to scan the entire database was a built in security flaw and one of the reasons (among others) bots were able to have their way with it. Be glad the new site only brings up one page at a time. Along the lines of what you are suggesting someone would have to navigate to each individual thread - and then ctrl-all to highlight the contents of the entire thread - and ctrl-v to copy the highlighted text onto the Windows Clipboard. From there it could be pasted/appended into an editor where it could be written to a .pdf file. Do that periodically for each thread to keep it current and you have it. Seems like a lot of work I know... But so is writing and implementing a search engine. < G > -jp .
| |
| Hdcper 12/28/2008 11:18:03 PM | Steve, What is wrong with the AFR rank? Bill
| |
| jeff 12/29/2008 12:18:59 AM | Let's continue this discussion in the Gap Rev Boxes thread over in the Private area of the site. But so demo users and would be customers can know what we're discussing (they can't see the private section of the site) here's the first post in the thread: In one of the program updates late last year or early this year I added two checkboxes to the System Definitions Screen. The boxes are shown in the screenshot below: http://www.jcapper.com/images/gaprevboxes.jpg The purpose of the boxes was to address a bug that Steve had found last fall. The bug had been there since the introduction of the program. The nature of the bug was that the program was basing factor gaps and ranks on all entries in a race including those that had scratched out. This meant that when a Any Factor rank=1 horse with a 2 pt gap over the second best horse in the field was scratched the new Factor rank=1 horse was still being identified as rank=2 with a gap = -2. Seems downright nasty right? It's not but read on. Even scarier to me was the fact that a couple of years of hard R&D into PMI, CMI, Opt Pts, etc. which eventually led me to JPR, JPRClass, and JPR ML and Tote Prob were all based on databases with the bug built into the numbers. Correcting the bug at my end and rebuilding all my databases from scratch didn't change my numbers significantly. In fact I got better JPR results with the bug in place than with the bug removed - which makes sense to me because it was there all along when I did the research that led me to JPR. So rather than force the bug fix onto everybody I created the two checkboxes that you see on the System Definitions Screen. The idea behind the boxes was to give the user a choice: A. Implement the bug fix. -or- B. Leave things alone. Note that there are two checkboxes - one for JPR and one for UPR. On my own Sys Defs Screen I have the JPR box unchecked. Which should make sense if you've read this far. It means I still want the bug introduced into the way my copy of the program calculates JPR. But notice that I have the UPR box checked. I developed my current UPR after publishing the program version with the two boxes. All of the R&D that led to my own UPR was done with the box checked. That should make perfect sense if you've read this far. Here's how the whole thing works: If neither box is checked the program acts as it has since day one with the bug intact. If either box is checked the program removes scratched horses during Calc Races and Database Build routines before writing ranks and gaps to export and playlist files. If only the JPR box is checked the program removes scratched horses during Calc Races and Database Build routines before writing ranks and gaps to export and playlist files. The program handles JPR with the bug removed. But, because the UPR box is not checked the program calculates UPR as if the bug were still there. If only the UPR box is checked the program removes scratched horses during Calc Races and Database Build routines before writing ranks and gaps to export and playlist files. The program handles UPR with the bug removed. But, because the JPR box is not checked the program calculates JPR as if the bug were still there. As posted on the old msg board at the time the version of the program was published that contained this bug fix: The best way to decide whether or not to check the boxes based on your own UDMs and UPR is to rebuild databases from scratch with the boxes unchecked and look at your results in the Data Window. Then rebuild from scratch with the boxes checked and look at your results in the Data Window. Note: You can preserve playlist files by renaming them to something other than pl_profile.txt prior to a rebuild from scratch. Just be sure to keep the first characters equal to "pl_" without the quotes. The differences to JPR and UPR with the boxes checked vs unchecked should not be significant. But the only way to tell whether or not to check the boxes is to rebuild some databases from scratch, take a peek in the Data Window - and see what looks better to you. Hope I've explained that in a way that makes sense. -jp . ~Edited by: jeff on: 12/29/2008 at: 12:18:59 AM~
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |