At 1/15/05 10:32 AM, Bahamut7 wrote:
I saw my new pace and I thought "Shit". From 36.?? to 25.14.
The Ba just crossed past 21 yesterday, so you're in good shape.
At 1/15/05 12:10 PM, Denvish wrote:
it's probably a stupid thing to say, but it would be nice if the Wi/Ht people could be somehow involved in the progression of NG as a site.
It's not stupid. gfox spoke his piece during his Pentalist update (please make a VP list), while I spoke mine during this update (please track changes). We may be better off simply starting up a thread to make it easy for James to track down all of our suggestions. We'd just need a mod to keep an eye on it to cut out the usual drivel that will find its way in, ala, "y duz peeps get 2 vote when dey hav no flash"
Thanks for the update. Any chance (don't feel obliged) that you could run another 'days to EGSC' at some point?
Sure, why not.
Also, did you get my latest email (about v2)? Hopefully that'll help things along a bit
Yes, but I haven't had a chance to test it out yet. Heck, they even got a ~36 hour respite from me forgetting to start up again after a reboot.
At 1/15/05 03:45 PM, EKRegulus wrote:
I don't like the way it sounds :(
I tried to cover the most likely situations, but there's bound to be a few more of each out there that do occur that I didn't mention.
I really doubt that there will be an official top blammers, top saviors, top posters (DUH) lists. Even an extended top reviewers list probably won't happen :)
The reason for all of these lists is that we can't see further down the rankings than top 10 or 50. If James is trying to eliminate our need for usage of the DM or PG, he will have to extend them. The pages are created at ~4:35am, with profiles being updated at 4:45am, so chances are pretty good that any expansion will simply continue that process rather than being dynamically created to reflect ongoing changes.
Top Posters list, although liked by some, is just asking for trouble, so the likelihood is that there won't be one, although it's still possible. I mentioned it simply because it's another of "the lists."
You're likely right that Top Blams/Saves lists won't be created, but if he's making a VP list, they could join the lists. Doing so should be a fairly easy process, but even if "by stat" lists show up, it's extremely unlikely that "by %" lists will come along for the ride.
That's a horrible fate. I really hope it turns out differently or at least that some of you find a way to continue some of your lists without being obliged to torture yourselves.
If there's a list expansion, what will end up happening is that pages will generate (similary to the EGRL, XP list, and TL) that simply show the ranks, names, and totals.
SK and Denvish could easily rewrite their programs to pull from them for ramagi or I, say data off 30 pages of 50 profiles instead of pulling 1500 pages, but the problem would arise due to the (relatively) dynamic nature of the pages. I could easily show the rate of change of the person ranked 600. The problem is that who is 600 now is not who was 600 last week.
Presuming the new NG lists are simply extensions of the information given now, my options to track changes in rank and totals are limited ... and none are good.
The simplest of all solutions is simply to drop it entirely. You folks won't like it because there's the lack of information. I don't like it, because it's a challenging problem I know I can solve.
The "old way" I had been doing things was to manually go through the cell referents fixing them to reflect the people, not the rank. At roughly an hour per 300 profiles, it's a pretty bad way of doing things.
Another option I have is to write out a series of formulae that would cause the spreadsheet to read the name, then compare it to everyone else on the list, spitting out the change when it finds a match. This would require one cell to be written for everyone on the list and that entire set of cells used with everyone on the list, plus a series of cells to consolidate that information. For just this past update, I would need ~7 million cells (~1.7m per variable plus organizational cells) compared to the 165,000 I used this week. If I don't mind having 4GB+ spreadsheets, it's not great but it's still viable.
The most feasible is probably to simply change the way the spreadsheet works and drag my way of thinking along with it. Rewriting the macros and cell codes to manage an alphabetized list, rather than one that works by rank, would be the best way to go. Technically, it wouldn't require much work to alter it, while also setting it to be able to post it based on numeric criteria, but recalibrating my head to think about it alphabetically is going to be an exercise in futility.
Lastly, my final option is to go out and buy (and learn) Access. Rewriting the spreadsheet like above is essentially just a way of setting up a relational database, keying the formulae to the names rather than the rank. The real problem with this is that I'm utterly devoid of experience with the program. Not only do I not know how to get it to do what I want, I don't even know that it can do what I want. Even if it can get things organized as I wish, can it put it into an easily postable format? There's just too many unknowns for me to take this way seriously.
Looks like we have a nice race for EGSC. :)
Yoink's indicated that he's trying to make 28k before school, so he'll likely "stall" somewhat and take a bit to get there. moongod, X-Naut, Syme, and good_guy_o are also in the hunt for SC as well. There's going to be a lot of big boosts to the top end of the VP list over the next couple of months as people hit General and SC =)
