00:16:30 * davb works on the top-secret project... 00:28:34 hey dave. 00:29:41 hmm. 00:30:36 hi. 00:30:44 Dave! 00:30:46 This is all graphic design work anyway. 00:30:56 wanna help me out with some logic? 00:32:03 Sure. 00:35:52 *cough* 00:36:40 ? 00:37:20 oh 00:37:21 hey 00:38:02 okay, i have pages stored in www/$cityname 00:38:10 and a database of postal codes. 00:38:43 ok. 00:39:00 when multipul cities are in one code, i need to display a list of links to different pages. 00:39:08 how would i store this list? 00:39:23 YOu need a city to postal code mapping table? 00:39:25 keeping in mind there could be over 1000 of them 00:39:39 I guess. 00:39:55 Do you have this information, or will you have to enter it manuallly? 00:40:13 well, I think i can get a script to enter it for me. 00:40:27 I will get it off a DB. 00:40:37 Ok. that is good. 00:40:57 for now my table is like this 00:41:11 postalcode # 00:41:11 void_p #if this code is currently unassigned. 00:41:12 muticity_p #if the code area contains more than one city, this is true. 00:41:12 city #The city this code is in. 00:42:09 In the US, there can be a city with multiple ZIP codes within it, or a zip code with more than one town or village. 00:42:12 so if multicity_p =t it needs to display the appropriate list of links. 00:42:24 SO I think you need 3 tables. 00:42:29 oh. 00:42:32 wonderful :) 00:42:47 hmm 00:42:53 Postal Codes, Cities, and a mapping table. that links each postal code, city pair. 00:43:04 make sense? 00:43:07 sort of 00:43:22 There is something resembling this I think in ACS isn;t there? 00:43:28 but... the only goal is to direct the person to the right city page. 00:43:42 I dunno davb, I'm newbie. 00:44:06 Ok. They will have a postal code, but not the city? 00:44:14 ? 00:44:33 I mean, if there is multipul codes in one city that doesn't matter to me. 00:44:42 Ok. 00:44:53 they all go to the same page... 00:44:58 Ah. 00:45:13 BUt it is possible there is more than one city in a postal code? 00:45:30 i belive so. 00:46:02 I just got my info through rumor tho. 00:46:07 I can't think of a normalized way to do it without the 3 tables. 00:46:09 :) 00:46:13 :) 00:46:40 hmm. 00:47:00 maybe there isn't codes with more than one city... 00:47:42 Right. you need to find out. if so, your system wil work, BUT you data could become inconsistent, because if there is more than one code in a city, the city will be repeated. 00:48:11 For this ZIP Code, ZIP Code 00:48:12 City Name State the city name is: Type 00:48:12 ---------------------------------------------------------------------- 00:48:12 PASADENA MD ACCEPTABLE (DEFAULT) STANDARD 00:48:12 LAKE SHORE MD ACCEPTABLE STANDARD 00:48:12 RIVIERA BEACH MD ACCEPTABLE STANDARD 00:48:35 see there are multipul here... 00:48:43 disclaimer: davb is not a database designer, he just plays one on IRC 00:48:56 haha 00:48:57 US is also tricky, they are all probably the same place with different names. 00:49:20 perhaps i should just use the default. 00:49:35 then it will work 00:50:20 sure. depends on your application. the 3 table design is the standard way to do this kind of thing I think though. You have a two-way many-to-many relationship or something like that. 00:50:25 all OACS has to do then is select the postal code and return the default city name into a variable... 00:50:58 Right, they all go to the same page? 00:51:07 yes 00:51:15 i mean... 00:51:28 they all go to $cityname.adp 00:52:43 so as long as each valid code returns the proper city name, we are happy. 00:52:53 Ok. 00:53:03 i like talking 00:53:07 that is an interesting way to handle it. 00:53:08 it helps me think 00:53:18 sure, that is what this channel is for :) 00:53:19 do you think it is foolish? 00:53:39 I think there might be a way to have only one adp/tcl page for all the cities. 00:53:52 hmmm. 00:53:58 like this /directory/cityname/ 00:54:07 wait... brain working... 00:54:17 maybe not. 00:54:31 ahhh 00:54:35 I see. 00:54:37 that is good. 00:54:39 there is the index.vuh 00:54:56 .vuh? 00:55:11 virtual URL handler 00:55:34 I am not sure if it can do what I am thinking though... 00:55:41 hmmm. 00:55:56 do you actually need a real filesystem directory for each city? 00:56:04 i dont think so. 00:56:10 that is the stage i am at now. 00:56:18 ok. 00:56:22 structure/interface planning. 00:57:13 but every page needs to have entirely unique content. however i think it can always follow a template. 00:57:32 so therefore i may need only one .adp 00:57:37 Right, it depends if they have the same template or not. if they are different, that might be a problem. 00:58:03 if they are the same, then you can probably use just one page that gets the city name and serves the corrent content. 00:58:05 I'm pretty sure the only logical asnwer is they need to follow a template 00:58:20 for realistic maintainance 00:58:26 right. 00:58:34 hmmm. 00:58:39 the OpenACS tempalting system is very powerful. 00:58:50 00:59:14 how is your deal going. with that one guy. 00:59:37 good, better if he actually gets some investors. 01:00:05 oh, you are at that stage already? or are you just looking for seed capitol? 01:00:32 seed capital. after we get a little demo to shop around. 01:00:56 thats cool. 01:01:46 *ponder* 01:02:45 there is only one paragraph explaining VUH that I can find :( 01:04:09 amasing documentation :) 01:04:30 yeah, they never finished, barely started in places.... 01:04:53 I think it will do what I think it will do :) 01:05:08 which is? 01:06:04 that somehow you can have one tcl file that will figure out which city is requested in the URL and serve it. so you don't need a seperate page or soemthing like ?city=cityname 01:06:24 the vuh system is supposed to replace registered procs so I should be possible. 01:06:31 *nod* 01:07:38 so for anonymous users, how does the OACS keep track of the users City? 01:07:55 cookie? 01:09:32 yes. 01:09:42 I think you would need to set your own city cookie. 01:09:57 hmm? 01:10:15 I don't think OpenACS cares about the city in the default install does it? 01:10:32 no. 01:10:34 I assume a visitor would be able to set the prefered city even if they don;t register? 01:10:40 right 01:10:46 http://www.arsdigita.com/bboard/q-and-a-fetch-msg?msg%5fid=000JTn&topic%5fid=21&topic=web%2fdb 01:10:47 A: http://www.arsdigita.com/bboard/q-and-a-fetch-msg?msg%5fid=000JTn&topic%5fid=21&topic=web%2fdb from davb 01:11:00 A:|Thread on Virtual URL handlers for ACS 4 01:11:00 titled item A 01:11:32 its all users, noone from aD choose to explain how it works :( 01:11:40 :) 01:12:11 hmmm. 01:13:06 it looks liek you need a folder /cities/ and a file in there index.vuh, then link to cities/cityname when the cityname file does not exist it reads index.vuh 01:13:06 I'm thinking, if i'm asking for a postal code already, perhaps i should ask for email address at the same time. 01:14:27 sure. maybe make it optional, but offer better service with it than without? 01:14:46 no 01:14:50 well kind of 01:15:07 offer the best service posible at all times 01:15:23 right. 01:15:26 but naturaly there will be better service if we have more info. 01:16:00 I still don't know if anonymous coward should be able to post 01:16:07 but i'm thinking yes 01:16:25 denshi (~chatzilla@adsl-216-62-223-193.dsl.rcsntx.swbell.net) has joined #openacs 01:16:33 I think so too, but you can have a quality issue. 01:16:40 at least until there is a large amount of data available, and the network hits a critical point. 01:16:47 So you probably need someone to look through them for appropriateness 01:16:49 hi denshi ! 01:16:58 yo davb 01:17:11 I'm thinking of stealing /.'s system 01:17:26 but making it more simple and appropriate 01:17:53 good idea. 01:18:31 i'm thinking only moderators can mod, but people frequently modded up will get an invitation to be a moderator. 01:19:22 also moderation would be invisable to the users. 01:19:44 Sure, that is what philg says in the book, but never programmed it in :) 01:19:54 haha 01:19:56 cool. 01:20:06 davb: I hear your supersecret project is going well. 01:20:36 Hopefully. 01:21:09 I really hope we can get something going and get a working site so I can show people :) 01:21:16 :) 01:21:18 me too. 01:22:33 I'm not sure I want to be using ad_returnredirect want because it redirects to a URL with the ?'s and &'s in it and search engines may ignore those pages. 01:22:38 is this accurate? 01:22:46 google ignores nothing! 01:23:04 how about the less fun search engines? 01:24:02 loggy, who is jim and why does he need to talk to me? 01:25:20 I'm logging. I found 1 answer for 'who is jim and why does he need to talk to me' 01:25:21 0) 2002-01-22 01:24:02 loggy, who is jim and why does he need to talk to me? 01:25:27 denshi: hmm, my xchat tab turned blue :) 01:25:27 * denshi praises loggy. 01:25:43 he works like grep :) 01:25:55 loggy, shut up and learn to parse. 01:26:12 haha 01:26:13 better. 01:26:51 loggy, can you find the stick for 'talli: ok, I'll talk to todd' 01:26:51 I'm logging. I don't understand 'can you find the stick for 'talli: ok, I'll talk to todd'', denshi. Try /msg loggy help 01:27:34 I hear you're doing an ns :) 01:27:35 ruby 01:27:53 yup. 01:28:00 I have a back^n burner project called perl-aol 01:28:45 Are you Jim Freeze? 01:29:10 does things, one, makes libperl available as an aolserver module... two, lets you run perl scripts... three, exposes some of aolserver api 01:29:13 no 01:29:18 * talli is proud of himself for the possible hacker shiduch 01:29:19 Jim Lynch 01:29:35 yup, that's what sf.net sez. 01:29:42 talli has quit ("See ya!") 01:30:04 what is a shiduch?? 01:30:31 I'll have to add talli to my list of "People who hold their knowledge of yiddish over my head." 01:31:05 oh, no releases since 0.0.3. 01:31:09 right now, perl-aol crashes aolserver... 01:31:27 I can't resolve a problem in my destructors 01:31:30 Yes, I tinkered with perl-aol before starting ns_ruby. Thanks. 01:31:40 np 01:31:47 you did? 01:31:59 beattiek: only google matters :) 01:32:08 davb: thanks :) 01:32:12 I certainly didn't comprehend it, but I did dig at it for clues. 01:32:14 why tho? 01:32:25 beattiek: in reality, I think all real search engines do not ignore ? anymore. 01:32:32 cool 01:32:49 so what is the advantage of .VUH 01:32:57 see, its first goal is to get a very efficient set of bindings of aolserver api calls into perl 01:33:24 beattiek: it makes pretty URLs 01:33:27 right, 01:33:30 no new conversions in this layer 01:33:46 davb: its the "official" OpenACS 4 replacement for registered procs. 01:33:51 Would it be difficult to switch to .vuh afterthe stage where you add content? 01:34:15 so is this something you plan to pick back up? 01:34:17 beattiek: actually, no. if you change URLs the vuh should be able to translate it to the new one. 01:34:31 yeah, except it never gets onto my plate 01:34:41 I need to actually try this out and report back :) 01:34:47 :) 01:34:56 vinod: denshi mentions Guiness. 01:35:34 jim: If you want my advice, and you probably don't, I'd suggest you drop it and work on parrot-aol. 01:35:47 what's parrot? 01:35:53 :) 01:35:58 I looked first at your work, then at the perl internals, then at embedding perl, then they locked me up. 01:36:10 heh 01:36:36 it's the reference counting that I don't quite understand yet 01:36:47 After I was sane again, I determined that the perl 5.x.y interp is *way* too hairy to embed, and extremely difficult to extend. 01:36:48 mmmmm guiness 01:37:05 vinod: it really is the best in Ireland. 01:37:23 i really have to make a trip 01:37:30 next oacs social in dublin, anyone? 01:37:53 nonoNO next one in san francisco 01:37:54 next oacs social in Cork! 01:38:03 no, wait, Blarney! 01:38:23 no, Killkenney! and then we can watch a Hurling match! 01:38:45 * vinod agrees to all of the above 01:38:53 jim, did you even come to the SF social? 01:38:54 they killed kenny! those bastards! 01:39:09 umm, I think so 01:39:33 at a hotel in Noe area? 01:39:44 Ben was there? 01:40:16 uhh.. I vaguely remember that one. I was asking about the one talli & I threw in Buena Vista. 01:40:35 no, dont' think so... when was that? 01:41:41 was there a signup for it? 01:42:35 anyway, jim, the perl docs scream bloody murder about perl extensions and just curl up into a little ball and weep about embedding perl. That's one of the primary impetuses in Perl 6, to redo the interp. The new interp is called Parrot, it is mostly the product of Dan Sugalski, and it is planned to be a useful runtime for any language that can write to it. Perl, Python, & Ruby are being worked on. 01:42:39 ah. we should have one at Dinosaur Bar-B-Q 01:42:47 what is that? 01:43:06 So, as I said, a parrot-aol would be much cooler, and much easier to work with. 01:43:44 http://www.dinosaurbarbque.com/ 01:43:44 B: http://www.dinosaurbarbque.com/ from davb 01:44:10 B:| Dinosaur Bar-b-que 01:44:10 titled item B 01:45:00 davb has drawn forth a great hunger in me.. 01:45:18 sorry...:) 01:45:32 vinod, are you in boston? 01:45:38 That is the best place to eat in new york state outside of NYC 01:45:42 davb: 01:45:50 i see. 01:46:03 Thats where you just were? 01:46:21 yes. in syracuse. I haven;t been to the rochester one yet. 01:46:44 I'll go there, next time i'm in NY. 01:47:01 denshi: yup 01:47:46 there must be a few hurling teams around there... are you familiar with this sport? 01:47:58 I suppose NY is the second best place in USA for seed capital isn't it? 01:48:19 hurling... nope, never heard of it 01:48:36 what perl needs to be embeddable, is an input stream setup that has some extra pieces for doing some types of backup... you give all this to perl and you shoud be able to say "run one perl statement, I want to see the next symbol after it so I can decide if the embedding is done" 01:49:05 you can view backup as lookahead too if you want 01:49:12 bbiaf 01:49:44 beattiek: I hope so :) 01:50:30 vinod: field hockey on drugs, as only the Irish could play it. 01:51:28 cool - i'm reading about it on hurling.net 01:51:33 * denshi can't find a copy of the NY philharmonic performing Duke Ellington's Straynote Symphony. 01:51:37 sounds like something i'd like to stay away from 01:52:53 heh. Sometime soon I need to go back to Eire and take a bicycle. 01:53:21 *that* sounds like fun 01:53:58 I had no idea when I went there (I was 17). It's the perfect bike zone. 02:02:08 okay, it's time to head down to ye local Beer Snob Pub. 02:05:35 denshi has quit () 02:23:30 vinod has quit ("changing universes") 02:32:56 mindst0rm (~mindstorm@ma-ppp159.mva.net) has joined #OpenACS 02:33:49 hi mindst0rm 02:34:49 hi. 02:36:31 I was looking at getting a test install of ACS Java up and running on a small site (4000 page views a day). What kind of hardware would I need for oracle 8i to run at a comfortable rate? 02:37:11 depends on definition of comfortable. I have run oracle for testing on an AMD K6 400 with 256 mb ram. 02:37:12 I think that 256MB of ram on any decent computer would do fine. 02:39:29 given that we do 1000 - 4000 page views a day. Would a dualie setup be overkill? Would it RAID 1+0 be worth considering. ACS Java will be running on a seperate box (HP-UX box 200mhz). 02:44:26 I would say you need first as much RAM as you can afford. Mirroring is good to insure your data. If you have enough RAM and are going mostly reads the disk performanace won't matter so much. 02:45:15 thank god RAM is still cheap right now. 02:46:02 I'm worried that our Pentium II BX chipset box can't handle > 512MB RAM> 02:53:29 has anyone tried ACS Java out on jboss? 02:54:14 not me. I doubt there are too many people who have tried ACS Java in this channel :) 02:55:14 right, but hopefully there will be a java version of OpenACS that runs on postgres :-) 02:56:00 interesting idea. 02:56:45 I'm not to fanatical about AOLServer or tcl. 02:57:04 neither am I, but it works :) 02:57:56 It does the job right, but I have to totally relearn my skillsets to implement. I've alot invested in Apache. 02:58:27 speaking of which, what's the deal with mod_tcl or mod_aolserver? 02:58:41 Noone is really interested I think. 02:59:08 If you are going to use TCL and teh AOLserver APIs you already learned all that you need, its easier to run AOLServer. 03:05:38 thanks guys 03:05:42 mindst0rm has left #OpenACS 03:06:47 vinod (~vinod@207-172-97-132.s132.apx1.sbo.ma.dialup.rcn.com) has joined #openacs 03:15:47 markd2 (~Snak@166.102.30.119) has joined #openacs 03:19:31 talli (talli@lti-4.dialup.access.net) has joined #openacs 03:26:13 HOLA 03:27:56 HOWDY 03:28:04 how goes it? 03:28:05 HI!!!!!!!!!!! 03:28:10 GOES GREAT!!1!!1!!! 03:28:10 did you slay the dragonfly? 03:28:16 no, it's still there 03:28:22 and... as I walked up the stairs into the wolfpad tonight 03:28:28 a lizard popped out 03:28:29 oh my, did it get any bigger? 03:28:38 so.. now we have a pet lizard in the house. 03:28:54 perhaps he is man enough to take on the dragonfly 03:29:09 i always knew that florida was a backwards jungle land 03:29:13 is it true that there was a man in here tonight trying to find info on acs-java? 03:29:34 he is using acs-java, but has oracle-related questions. 03:29:35 * vinod looks shocked 03:29:52 vinod, i just came back from an elegant night of Harness Racing.. .and you dare call this place backwards?? 03:29:57 poor man reads like he is doomed 03:30:05 yeah. 03:30:20 he is hoping for ACS Java working with Postgresql. 03:30:32 wish him luck. 03:30:54 he's doomed 03:31:38 too bad markd2 wasn't here to set him straight 03:31:49 docwolf: sorry, i take back my hasty comment 03:31:50 my bad 03:32:05 harness racing? sounds kinky 03:32:07 I tried to show him the difference between ACS Java and OpenACS 03:32:18 it was "classy". 03:32:22 one is a giant hulking mass of code 03:32:24 the other is written in java 03:32:26 (in a 50s sinatra kind of way) 03:32:32 right 03:32:39 ... total time-warp experience. rolf will never be the same. 03:38:33 (of course, harness racing is only the beginning... i must now introduce him to the true "sport of kings" -- jai alai. 03:52:59 davb has quit ("ChatZilla 0.8.5 [Mozilla rv:0.9.7/20011221]") 05:20:19 vinod has quit ("changing universes") 06:41:00 dlk (dlk@garrison.ita.chalmers.se) has joined #openacs 06:49:31 djg has quit (Read error: 60 (Operation timed out)) 08:18:10 markd2 has quit (Read error: 113 (No route to host)) 09:36:53 djg (~dirk@212.84.246.68) has joined #openacs 10:02:11 dlk is now known as dlk-feeding 10:16:01 dlk-feeding is now known as dlk 11:17:26 dlk is now known as dlk--away-at-library 11:49:46 dlk--away-at-library is now known as dlk 11:53:18 vinod (~vinod@207-172-255-22.s1038.apx1.sbo.ma.dialup.rcn.com) has joined #openacs 13:32:46 markd2 (~Snak@r-41.124.alltel.net) has joined #openacs 13:32:47 vinod has quit (Read error: 104 (Connection reset by peer)) 13:35:15 davb (~dave@cm-208-136-23-203.nycap.rr.com) has joined #openacs 13:44:51 http://www.semanticstudios.com/publications/semantics/innovation.html 13:44:52 C: http://www.semanticstudios.com/publications/semantics/innovation.html from davb 13:44:59 C:|Innovation Architecture 13:44:59 titled item C 13:45:09 C: moving towards self-organizing web sites? 13:45:09 commented item C 13:51:28 * markd2 prefers web site entropy 13:55:00 like we have now? 13:55:05 yep 13:55:11 and we can contribute to that 13:55:22 by introducing self-disorganizing web sites 13:55:29 I see. 13:55:32 interesting idea. 13:55:41 better living through chaos 14:14:38 don't all the new ideas say that choas is self-organizing? :) 14:15:55 uhoh... people are googling for my name... or at least someone with the same name as me. 14:16:02 heh 14:16:16 docwolf is down to #5 for search string in my reports. 14:21:09 http://www.webtechniques.com/archives/2002/02/desirevu/ 14:21:10 D: http://www.webtechniques.com/archives/2002/02/desirevu/ from davb 14:21:18 D:|Adobe gets into CMS 14:21:18 titled item D 14:21:23 D: Server-side photoshop? 14:21:24 commented item D 14:21:31 D: haven't they ever heard of ImageMagick? 14:21:32 commented item D 14:22:00 D: Yahoo store can also custom render graphics, and has been for quite a while 14:22:00 commented item D 14:25:25 heh 14:25:57 its $7500 14:26:18 I am pretty sure I can whip up a few scripts to do this for half that :) 14:28:48 http://mpt.phrasewise.com/2002/01/21 14:28:48 E: http://mpt.phrasewise.com/2002/01/21 from davb 14:29:04 E:|Mozilla tool to support Blogger API 14:29:04 titled item E 14:29:19 E: wow. hook this into OpenACS/Edit-this-page 14:29:19 commented item E 14:29:37 E: Mozilla may be the desktop application to integrate with your CMS 14:29:37 commented item E 14:30:32 hmmm. if I modified chump to begin his messages with # instead of - they would automagically not be logged by loggy. 14:35:10 loggy could be the chump backup 14:35:33 true. 14:35:49 I am going to experiement with xena/thecreature. It does everything. weblog, log, google, etc... 14:36:21 maybe I will hook it into postgresql or aolserver or something. 14:40:32