| 00:02:01 | SpaceCoaster | Lenolium: hi |
| 00:28:57 | Lenolium | *plong* |
| 00:34:23 | other | will linux work right if i upgraded the ram in the tuxscreen? |
| 00:34:28 | other | [ie use all the mem] |
| 00:46:09 | as2 | I dont think lack of ram is the problem. |
| 00:47:05 | as2 | I dont think upgrading the ram is even feasable anyway. |
| 00:50:56 | other | why? |
| 01:00:03 | GoRK | Linux works only when there are 8MB onboard or 8MB onboard + 8MB SODIMM. If blob is fixed, it will be possible to at least support 8MB onboard + 32MB SODIMM |
| 01:00:03 | GoRK | Inferno will operate with up to a 32MB SODIMM along with the 8MB onboard |
| 01:00:57 | GoRK | spacecoaster: any luck? :) |
| 01:02:48 | other | <Q> spacecoaster No such nick/channel |
| 01:02:49 | other | :) |
| 01:03:17 | GoRK | hrm didnt see that he had left |
| 01:16:53 | GoRK | ok im getting the response from the dsp when i go off hook that line 1 is in use and no dialtone was detected in 2.1 second period |
| 01:16:56 | GoRK | then a sanity response |
| 01:21:35 | GoRK | aww crap.. RELAY LED is not a light but a relay that connects the phone line up.. which is why when it gets taken off hook you can still sort of hear it.. makes sense if it's an inductive relay |
| 01:21:44 | GoRK | !!! let us try this hypothesis |
| 01:28:52 | GoRK | that's it |
| 01:29:02 | GoRK | RELAY LED attaches the phone line to the dsp |
| 01:33:24 | GoRK | PARALLEL LED might have something to do with it too |
| 01:36:23 | GoRK | i should write some code |
| 01:53:40 | GPSFan | BZFlag: Hi, I'm back, and have found an eps to pdf converter. I converted all the schematics and put them up on my ftp site. Take a look, I think it's the same converter Tangent used, so the issues are the same. I've also put the postscript files up, they end in .prn but are really postscript. Just send them to a PS printer and you will have an A size print. |
| 01:54:00 | GPSFan | BZFlag: the url is: ftp://www.desertweyr.com/pub/TuxScreen/ |
| 01:55:57 | BZFlag | you could rename them to .ps ;-) |
| 01:57:19 | GPSFan | Yep I could have , but why should I have all the fun? |
| 01:57:45 | BZFlag | well, cause then I would just be able to click on them in netscape and they would come up in ghostview. ;-) |
| 01:57:57 | GPSFan | It probably took me more keystrokes to explain it that it to do it , I will asap. |
| 01:58:19 | GPSFan | standby.... |
| 01:59:35 | BZFlag | do you have a sf account? I can get you upload permissions to TuxScreen.net |
| 01:59:41 | GPSFan | done |
| 02:00:02 | GPSFan | Nope, that would be nice, thanks |
| 02:02:53 | GPSFan | They should be .ps now |
| 02:25:02 | GPSFan | Now they are scaled properly, so they aren't all scrunched to the top of the .pdf. |
| 02:30:58 | GPSFan | BZFlag: I now have an SF account. |
| 02:33:41 | BZFlag | nice. what's your SF account name |
| 02:34:15 | BZFlag | er, that was a ? ;-) |
| 02:37:45 | GPSFan | kenebm |
| 02:40:44 | GPSFan | BZFlag: sorry, I'm trying to multiprocess here, ring out more of the Tux, enter schematic data, watch tv, scratch teh cats, etc.etc.etc.... |
| 02:40:59 | GPSFan | s/teh/the |
| 02:47:37 | SpaceCoaster | hi |
| 02:47:51 | GPSFan | SpaceCoaster: hi |
| 02:48:20 | BZFlag | GPSFan: you now have access. |
| 02:49:02 | GPSFan | Thanks, BZFlag: I figured out how to use the translator program, and the .pdf's are much better now. |
| 02:51:45 | BZFlag | cool, I'l check it out in a minute. I moved the schematics directory over to be under download. |
| 02:51:56 | BZFlag | feel free to drop new stuff in there. |
| 02:53:11 | GPSFan | BZFlag: make sure you moved it after 21:27MDT since those are the properly scaled ones. |
| 02:54:15 | BZFlag | there aren't any there now. just the cave drawings. |
| 02:54:50 | BZFlag | the cave drawings used to be on /img/schematics I just moved them to /download/schematics/ |
| 03:24:15 | sammy_wms | morning all... |
| 03:24:48 | as_ | Evnin sammy. |
| 03:25:44 | GPSFan | BZFlag: I must be dense, but where on sf can I find out how to upload to the schematics directory? |
| 03:32:06 | BZFlag | scp this.file kenedn@tuxscreen.net:/home/group/t/tu/tuxscreen/htdocs/download/schematics/ |
| 03:33:41 | GPSFan | BZFlag: how about kenebm..... |
| 03:35:03 | BZFlag | oh, yeah. sorry. |
| 03:35:06 | BZFlag | got scp? |
| 03:35:13 | GPSFan | yep. |
| 03:36:06 | BZFlag | oh, and it's group not groups. |
| 03:38:13 | BZFlag | scp this.file kenebn@tuxscreen.net:/home/groups/t/tu/tuxscreen/htdocs/download/schematics/ |
| 03:38:23 | BZFlag | er, groups not group I meant. |
| 03:38:41 | GPSFan | BZFlag: I think I've found it. |
| 03:40:22 | BZFlag | yep ;-) there they are! |
| 03:40:25 | GPSFan | BZFlag: all uploaded, check them out... |
| 03:41:44 | collord | BZFlag: I ordered =) |
| 03:42:18 | GPSFan | BZFlag: this is all new to me, I'm just a hardware guy, I've known of sf for a long time but never had the occasion to use it, it's pretty cool. |
| 03:43:29 | SpaceCoaster | GPSFan: nice drawings, much better size |
| 03:45:11 | GPSFan | SpaceCoaster: yea, it took a bit of playing with the converter, and the print parameters of Orcad, but I finally got it so both .pdf & .ps are ok. |
| 03:45:51 | GPSFan | SpaceCoaster: took all evening though, time for bed Zzzzzzzzz. |
| 03:46:38 | SpaceCoaster | me too, see ya |
| 03:46:55 | GPSFan | bye! |
| 03:55:19 | BZFlag | collord: got it, thanx. |
| 05:56:49 | GoRK | returns |
| 05:57:45 | GoRK | whoot |
| 05:57:47 | GoRK | anyone still up |
| 05:57:52 | Proteus | Yep. I'm here. |
| 05:58:14 | GoRK | sweet. i almost have this phone picking up and dialing a number with dtmf |
| 05:58:27 | Proteus | On the TuxScreen? |
| 05:58:32 | GoRK | yep |
| 05:59:00 | GoRK | one of the LED switches does not really control an LED. it controls a relay |
| 05:59:17 | Proteus | Thought so. :-) |
| 05:59:26 | Proteus | I was looking through those schems they scanned. |
| 06:00:29 | GoRK | anyway it seems to be a dpdt relay that bridges the phone line to the dsp |
| 06:00:44 | Proteus | That's usually how most modems do it. :-) |
| 06:01:39 | GoRK | anyway the only problem now is what sequence to do all the steps for the switchhook event |
| 06:02:16 | GoRK | since throwing the relay at the wrong time gets interpreted as a switchhook event back to the telco.. and then no more dtmf dialing |
| 06:05:09 | Proteus | *nod* |
| 06:05:15 | Proteus | That can be tricky. |
| 06:07:28 | GoRK | this LED command in cosmicpenguin's document must be wrong |
| 06:46:40 | GoRK | whoot got it working |
| 06:47:38 | Proteus | ?? |
| 06:47:51 | GoRK | the phone picks up the line and dials out |
| 06:48:24 | GoRK | ./buzby a3 44 00 00 03 a7 02 0xa6 02 58 0e and then you add 58 0X where X is the digit for each digit you want it to dial |
| 06:48:46 | GoRK | err the 0xa6 could just be replaced with a6 |
| 06:48:54 | Proteus | Schweet. |
| 06:54:52 | GoRK | ./buzby a3 44 00 00 0f a7 02 22 1a a6 02 58 0e 58 <-- better version |
| 06:55:15 | GoRK | enables dialtone detection and turns on line in use led |
| 06:55:24 | Proteus | Scweet. |
| 06:56:21 | Proteus | Very cool. |
| 06:58:16 | GoRK | apparently something has to keep talking to the dsp for it to stay open |
| 06:58:44 | GoRK | i assume that is so that in an emergency if the software freezes or something or for when the power dies, the phone continues to operate |
| 06:58:55 | GoRK | finally i got the LED's figured out :) |
| 06:59:14 | Proteus | Cool, very cool. I wish I had my box here so I could test it. |
| 07:00:45 | GoRK | this dsp seems to have controls for 7 unattached digital switches |
| 07:01:23 | Proteus | Maybe for PABX control. |
| 07:01:45 | BZFlag | there might be separate switches for the "aux" port? |
| 07:01:51 | GoRK | well it appears that they are in there to support multiline phones and the like |
| 07:01:54 | BZFlag | might be able to be a 2 line phne? |
| 07:02:11 | Proteus | You never know. |
| 07:02:24 | BZFlag | try dialing with the line on aux jack |
| 07:02:47 | Proteus | If I had a TuxScreen, I would. :-) |
| 07:02:49 | Proteus | I'm waiting on mine. |
| 07:02:50 | GoRK | BZFlag: i doubt it, the traces on the board are wired together |
| 07:02:53 | Proteus | Doh. |
| 07:02:58 | Proteus | Window switching is killing me. :-) |
| 07:04:16 | GoRK | BZFlag: but that doesnt rule out the possibility that the other jack could be reconnected somewhere else and it work as a 2 line phone :) |
| 07:05:47 | GoRK | a 2 line phone makes sense because you'd need an extra line indicator, an extra message light, an extra relay for phone, an extra relay for a bridging circut for conferencing (unless the dsp did it).. |
| 07:05:57 | GoRK | so that accounts for the missing bits |
| 07:06:38 | GoRK | in the led control |
| 07:08:38 | GoRK | hmm the flash rate has nothing to do with anything either :) |
| 07:08:52 | BZFlag | GoRK: but you are sure that's not the way it's wired now? |
| 07:09:17 | GoRK | BZFlag: i have a commboard right here. they are connected as soon as they hit the board |
| 07:09:35 | GoRK | on the back side |
| 07:09:49 | BZFlag | I had hoped to be able to build a voip gateway and digital answering machine on the same box, this would imply that aux can not talk to voip without being on the pots line (if attached) |
| 07:10:14 | BZFlag | ah, well. add a relay and cut the traces. ;-) |
| 07:10:21 | GoRK | there might as well be only one jack |
| 07:10:38 | GoRK | i think line2 could be wired up, sure |
| 07:10:58 | GoRK | but we'll have to wait for GPSFan to do his magic on that. I dont want to follow those buggers or even take my mongoose off the board |
| 07:11:46 | GoRK | the other problem is that we'll probably have to wire something up to the ucb1200 to get audio.. i'm thinking line2.. or maybe i could use the other side of that relay.... |
| 07:11:58 | BZFlag | yep. the 2 jacks only have 2 wires each and they are connected on the io board. bummer. |
| 07:12:49 | GoRK | hmm that's not a bad idea... dsp line1 interface normally connected to ucb1200 unless you flip the relay and then you're on pots |
| 07:13:37 | GoRK | could maybe go into the ucb1200's telecom interface directly and still leave ucb1200 audio ports open... possibly could easily install line in/out jacks on the side for connection to a stereo or something |
| 07:13:40 | BZFlag | hmmm.. why would the dsp want to talk to the ucb1200? |
| 07:14:04 | BZFlag | now, that would be nice. |
| 07:14:18 | BZFlag | ucb1200 to stereo jacks. ;-) |
| 07:14:43 | GoRK | the dsp doesnt want to talk directly to the ucb1200. the dsp just routes the audio data out to somewhere we can get it back in the system.. the most promising place to do that is in ucb1200 |
| 07:15:28 | GoRK | since the ucb1200 has 2 audio interfaces on it... a line level mic/speaker connections and a telecom interface suitable for connecting directly to pots line |
| 07:16:02 | GoRK | and has the added benefit of an 8 bit data bus connecting it back to the sa1100 vs the dsp's 1 bit nibbling POS |
| 07:16:10 | GoRK | which audio is never going over |
| 07:17:40 | GoRK | anyway cosmicpenguin's document needs serious revision :) the acutal interface does not seem to work exactly as documented (does it ever?) plus, there are some nifty hidden tones i found |
| 07:18:01 | Proteus | 4 nifty hidden tones? |
| 07:18:16 | GoRK | one would be quite suitable as a burgler alarm sort of thing. it's about the most obnoxious sound in the universe |
| 07:18:59 | Proteus | Heheh. |
| 07:20:50 | GoRK | anyway, i am now confident that we will not have to reprogram the DSP to do audio, but we will have to do a small hardware mod |
| 07:21:13 | GoRK | hopefully it wont require taking mongoose off |
| 07:21:51 | GoRK | BZFlag: do you know if the 5v pcmcia capable phone that shaggy has also had the dsp mod for voip? |
| 07:23:41 | BZFlag | don't know, but I presume so, |
| 07:24:05 | GoRK | any chance he'd send us some photos of the innards or something? |
| 07:24:21 | BZFlag | I asked, but "No Reply at All" ;-/ |
| 07:24:48 | GoRK | cool |
| 07:56:44 | GoRK | goodnight folks, i just posted my results investigating wheaties to the list for those curious people. |
| 08:54:19 | Ar3s | Salut |
| 14:21:26 | SpaceCoaster | Anyone else notice the Potty 'N Step Stool in http://TuxScreen.net/img/basement1.jpg ? |
| 14:22:00 | SpaceCoaster | Gues you have to make sure you close the lid first! |
| 14:22:45 | SpaceCoaster | Strange to see that combination of potty and stool! |
| 14:23:01 | SpaceCoaster | Is it a potty is it a stool? |
| 14:23:28 | SpaceCoaster | Wedge has an anal fixation, he would know. |
| 14:24:12 | SpaceCoaster | What happened to wedge aka wej did he go to wireless card heaven? |
| 14:26:11 | mallum | hi SpaceCoaster |
| 14:26:23 | mallum | I just built my box with the cats board |
| 14:26:55 | mallum | it wont boot the linux install cd though :( |
| 14:27:02 | mallum | boots the netbsd one though |
| 14:28:23 | SpaceCoaster | remind me, what's the cats board? |
| 14:28:43 | mallum | a strongArm sa110 board |
| 14:29:00 | SpaceCoaster | cool |
| 14:29:01 | mallum | thats atx and has pci and isa |
| 14:29:31 | SpaceCoaster | very cool, sortof an arm pc? |
| 14:30:02 | mallum | yeah |
| 14:30:16 | mallum | its in a crap pc case I got this morning TCR |
| 14:30:26 | mallum | s/TCR/from TCR computer fair/ |
| 14:31:07 | SpaceCoaster | strange an arm in a big box, what are you going to do with it? |
| 14:33:21 | SilverBullet | mallum: wrote buzby for you to control the DSP - have you had a chance to play? |
| 14:34:38 | mallum | SpaceCoaster: its gonna be a compile box, save me from x-compiling pain - its got 256meg so should work good |
| 14:34:55 | mallum | SilverBullet: yeah just about to check the post ! |
| 14:36:12 | SpaceCoaster | mallum: I've found the x-compiling quite friendly, haven't pushed it though, probably bad for debugging, if you have bugs that is. |
| 14:36:59 | SilverBullet | SpaceCoaster: lol |
| 14:37:32 | mallum | SpaceCoaster: yeah but somethings ( Python, gtk ) won't cross compile nicely as there configure scripts barf on not being able to run tests |
| 14:38:48 | SpaceCoaster | I understand, if they aren't setup for x-compile and they compile utils/tests then that would break. |
| 14:39:29 | SpaceCoaster | mallum: GoRK mailed some good stuff about command sequences which do neat things to the DSP. |
| 14:39:56 | mallum | SpaceCoaster: yeah they look cool, hopefully I'll get a chance to play |
| 14:40:30 | SpaceCoaster | You named the utility, buzby, so you better! |
| 14:41:40 | SpaceCoaster | mallum: do you know anything about enabling an interrupt in an Arm driver? |
| 14:44:21 | mallum | SpaceCoaster: heh, dont worry I'll play |
| 14:45:15 | SpaceCoaster | mallum: funny no ones asked why it's called buzby |
| 14:45:32 | mallum | heh |
| 14:45:43 | mallum | one sec, i gonna take a break .. brb |
| 16:07:14 | GoRK | hello |
| 16:11:14 | SilverBullet | GoRK: nice work with the DSP - guess you were up pretty late, SpaceCoaster and I just got RTS interrupts to work. |
| 16:11:26 | GoRK | cool |
| 16:11:37 | GoRK | i have a speakerphone dialing string if you want it |
| 16:11:40 | ctr | is the DSP progressing a fast rate? |
| 16:11:54 | GoRK | ./buzby a3 44 00 00 0f a7 06 22 1a a6 04 58 0e |
| 16:12:37 | GoRK | SilverBullet: can i send a command with buzby to the dsp to return the information i see in /proc/device/wheaties/spi_read ? |
| 16:13:08 | GoRK | SilverBuller: or can i just read bytes from /dev/wheaties to get it... the reason im asking is that im trying to figure out how to send it 'idle' commands |
| 16:13:16 | GoRK | or null commands or something |
| 16:13:34 | SilverBullet | GoRK: nice - thanks. |
| 16:14:03 | SpaceCoaster | GoRK: connects the line to the speaker but doesn't dial |
| 16:14:19 | GoRK | SpaceCoaster: you have to give it a number to dial after that |
| 16:14:40 | GoRK | put 58 5 58 5 58 5 58 1 58 2 58 1 58 2 to dial 5551212 |
| 16:15:07 | SpaceCoaster | GoRK: thanks, must learn wheaties one day :-) |
| 16:15:14 | GoRK | making the keypad work will involve a daemon |
| 16:16:20 | SpaceCoaster | wotz 5551212? |
| 16:16:33 | GoRK | information |
| 16:17:31 | SpaceCoaster | gargh they charge 2bucks for that, cheaper to dial the uk! |
| 16:17:53 | GoRK | heh. just replace it with something! did you dial it? |
| 16:18:22 | SpaceCoaster | yes, but i didn't let it connect |
| 16:19:33 | SpaceCoaster | GoRK: yeah just open /dev/wheaties, reads will return immediately if there is nothing to read. |
| 16:20:20 | GoRK | SpaceCoaster: cool. any ideas if sending the dsp something like 00 will keep it from timing out. it seems to have a watchdog timer in it |
| 16:20:33 | GoRK | timing out? <-- what i should have typed |
| 16:20:42 | SpaceCoaster | while(1) { n=read(wheaties_fd, data, 1024); sleep(1); } |
| 16:20:59 | GoRK | because reading causes a write to happen? |
| 16:21:59 | SpaceCoaster | the watchdog makes sense, they were probably required to ensure that this could dial 911 regardless of what inferno was up to. |
| 16:22:51 | SpaceCoaster | reading does cause a write of 00 for every byte read. |
| 16:23:20 | SpaceCoaster | We think you may have discovered why we need to reset on open. |
| 16:23:33 | GoRK | the watchdog? |
| 16:23:39 | SpaceCoaster | yup |
| 16:24:35 | SpaceCoaster | when interrupts are working and read data queuing we can try a keepalive timer in the kernel to ensure that something is sent every second. |
| 16:24:47 | GoRK | i am testing now whether reads are enough to keep this thing open |
| 16:25:19 | SpaceCoaster | hmm, if there is nothing to read (RTS high) then nothing will be sent. |
| 16:26:51 | GoRK | SpaceCoaster: reads are not enough to keep it from timing out |
| 16:27:34 | mallum | fails on Cats install :( |
| 16:27:51 | GoRK | SpaceCoaster: but i do get the good data |
| 16:29:08 | SpaceCoaster | try writing 00 |
| 16:29:19 | GoRK | 00 is a bad command. it returns 84 |
| 16:29:41 | GoRK | it'd probably keep it open but it's not the right way to do it... maybe i need to issue DSP_SANITY_COMMAND |
| 16:32:46 | GoRK | whoops |
| 16:32:47 | GoRK | heh |
| 16:33:10 | SpaceCoaster | was that an ooops or a whoops? |
| 16:33:26 | GoRK | a whoops |
| 16:33:42 | GoRK | i compiled up buzby with x86 gcc |
| 16:34:13 | AlHaz | heh |
| 16:34:59 | SpaceCoaster | mallum: you were right, this x-compiling can be tricky. |
| 16:35:08 | AlHaz | it sounds like we're going to need a kernel daemon to keep the wheaties happy, and possibly to facilitate touch tone dialing |
| 16:35:51 | SpaceCoaster | hmm, don't know about touch tone dialing, we could do that in userland. |
| 16:36:08 | AlHaz | *nod* |
| 16:36:19 | AlHaz | I know it's prefered to keep as much as possible in userland |
| 16:37:09 | SpaceCoaster | AlHaz: then we could keep logs, add shortcuts, etc. The people that designed this thing didn't put in a 10c chip to do DTMF they must have had a reason |
| 16:38:39 | GoRK | kernel daemon is overkill imo. it's easy to implement this all in userland |
| 16:38:50 | SpaceCoaster | AlHaz: we could also use the keypad as a replacement for the QWERT keyboard, just like a cellphone. |
| 16:38:57 | SpaceCoaster | GoRK: even the keep alives? |
| 16:39:32 | GoRK | SpaceCoaster: dunno why not.. the software watchdog timer in linux runs in userland |
| 16:40:34 | SpaceCoaster | ok, just if it dies the driver will have to recover so it might be more robust in the driver. |
| 16:40:52 | GoRK | the driver could do a sanity check every time there is a read |
| 16:40:55 | mallum | SpaceCoaster: heh, well when I get this box up ... I'll maybe give you an account ... maybe :) |
| 16:41:38 | mallum | SpaceCoaster: looks like I gotta boot it over tftp - except it dont pick up the damn network card ( obviously not ne2k compat ) |
| 16:42:02 | SpaceCoaster | thanks mallum, that would really be cool, but it was GoRK having the problems! |
| 16:42:31 | mallum | SpaceCoaster: heh ok GoRK too |
| 16:42:59 | SpaceCoaster | mallum: can we boot the TuxScreen over tftp I'm begining to worry about my flash, must have burnt 200 kernels for the touchscreen and DSP |
| 16:43:28 | SpaceCoaster | mallum: also bored watching the dots....... |
| 16:44:00 | mallum | SpaceCoaster: hmm not sure - have to check the docs on the blob |
| 16:45:21 | SpaceCoaster | mallum: might be tricky as it would have to init the ethernet card. |
| 16:45:31 | mallum | SpaceCoaster: yep |
| 16:45:56 | AlHaz | SpaceCoaster: you don't really need to worry about flash all that much |
| 16:46:00 | SpaceCoaster | watches the dots, again |
| 16:46:16 | AlHaz | SpaceCoaster: I've never spoken to anyone who's worn out a flash device who wasn't specifically trying to |
| 16:46:50 | AlHaz | should be able to erase & reprogram at least 100,000 times, usually a lot more |
| 16:47:40 | SpaceCoaster | thanks, I looked and they say 100,000 but they don't specify if that is per block or overall, it's not clear |
| 16:48:22 | AlHaz | SpaceCoaster: it's per block |
| 16:48:47 | AlHaz | SpaceCoaster: and it's just a minimum rating, same way cpu's are specced at a particular clock |
| 16:49:01 | mallum | hmmm, anybody know what the kernel oprion is for it to support RARP ? |
| 16:49:03 | AlHaz | YAMMV |
| 16:49:25 | AlHaz | mallum: oh, you have to enable boot time auto configuration under network support |
| 16:49:30 | AlHaz | then it'll show up |
| 16:49:59 | mallum | AlHaz: YAMMV ? |
| 16:50:10 | LIM | so is linux actually working at all on this thing yet? |
| 16:50:15 | AlHaz | Your Actual Mileage May Vary |
| 16:50:26 | AlHaz | (standard disclaimer in car adverts) |
| 16:50:38 | mallum | AlHaz: oh ok. is that option for servers too ? |
| 16:50:47 | SpaceCoaster | AlHaz: shucks I was trying to convince myself to buy another to share the load :-( |
| 16:50:48 | AlHaz | mallum: OH, you mean rarpd? |
| 16:51:03 | AlHaz | mallum: 2.4 doesn't have built in rarp server support, you need netkit-rarpd instead |
| 16:51:23 | AlHaz | SpaceCoaster: heh, well, you really should have a spare just incase. Never know, lightning could strike |
| 16:51:25 | mallum | AlHaz: a patch ? |
| 16:51:35 | LIM | a daemon |
| 16:51:40 | AlHaz | mallum: no, rarpd has to run in userland in 2.4 |
| 16:51:58 | AlHaz | thought you were asking about rarp client support |
| 16:52:03 | AlHaz | which is still built in |
| 16:52:11 | SpaceCoaster | AlHaz: yeah lost another port on the ethernet hub this week, 3 down, 5 left |
| 16:52:20 | mallum | AlHaz: hmm, when I try and run /sin/rarp - I get 'This kernel does not support RARP.' |
| 16:52:50 | AlHaz | mallum: Oh, hmm. |
| 16:53:11 | AlHaz | oh, right, no, that's correct with 2.4 kernels |
| 16:53:21 | AlHaz | in 2.2 kernels there is a rarp daemon in the kernel, but not in 2.4 |
| 16:53:24 | AlHaz | hold on . . |
| 16:53:35 | AlHaz | SpaceCoaster: ouch |
| 16:53:39 | GoRK | a kernel darmon makes it considerably harder for people who would want to change the behavior of the phone to do so... especially when you need to change the behavior on the fly |
| 16:53:48 | GoRK | a kernel daemon for wheaties, that is |
| 16:54:06 | AlHaz | http://packages.debian.org/testing/net/rarpd.html |
| 16:54:25 | mallum | AlHaz: yep got it |
| 16:55:21 | SpaceCoaster | GoRK: true, it was just for the keepalives, maybe it should be in userland though, so we use the same fallback facility as the original software. Don't want to be sued by someone whose house burnt down because they couldn't dial 911. |
| 16:55:53 | mallum | AlHaz: cool, now I just gotta get a ne2k compat card - am I correct in assuming all 2com ones arn't ? |
| 16:55:57 | mallum | s/2/3/ |
| 16:56:03 | AlHaz | yeah, 3com are their own deal |
| 16:56:08 | AlHaz | PCI or isa? |
| 16:56:14 | mallum | AlHaz: doh, all I have is 3coms |
| 16:56:18 | mallum | AlHaz: PCI |
| 16:56:33 | mallum | wonders what card his G4 has in it |
| 16:58:01 | AlHaz | mallum: look for something in a realtek RTL8029 |
| 16:58:35 | mallum | AlHaz: cool thanks for your help - dnner time now |
| 16:58:39 | mallum | is away: food |
| 16:58:43 | AlHaz | mallum: the chip will have a picture of a sand crab on it. don't get the 8129 or 8139, those are something else |
| 16:58:51 | AlHaz | should be dirt cheap |
| 16:59:48 | SpaceCoaster | AlHaz: wot are the odds of booting the TuxScreen over ethernet? |
| 17:01:08 | AlHaz | SpaceCoaster: Pretty good |
| 17:01:20 | AlHaz | SpaceCoaster: people are working on etherboot support for blob already |
| 17:01:50 | AlHaz | I've actually done it, but the patch was extremely ugly and has basically been thrown out |
| 17:01:56 | SpaceCoaster | AlHaz: I meant in the next 5 minutes :-) |
| 17:02:32 | AlHaz | heh, not good :) |
| 17:03:36 | AlHaz | I've seen it done, basically all they did was lift the ne2k boot stuff from the etherboot project and integrate it into blob |
| 17:05:35 | SpaceCoaster | going to look a blob next for the 32mb memory issue, don't want to get into networking and pcmcia at all, so I'll wait, thanks. |
| 17:07:26 | AlHaz | yeah |
| 17:07:56 | AlHaz | What I'm betting is that we can address the 32 meg sodimm as 4 banks of 8 megs. I'm just going on vague assumptions, tho. |
| 17:09:04 | SpaceCoaster | please share your vague assumptions right or wrong doesn't matter at this stage. |
| 17:11:46 | AlHaz | Well, someone was pointing out the other day that the sa1100 datasheet says that if there are multiple banks of memory, they must all be the same size |
| 17:12:02 | AlHaz | and currently the blob sets up four banks of 4 megs each |
| 17:12:40 | AlHaz | and with that setup, we don't seem to see a lot more banks of 4 with the 32 meg sodimm |
| 17:13:07 | AlHaz | but for whatever reason we're not treating the 8 meg on board and 8 meg on the sodimm as 1 bank of 8 each |
| 17:13:12 | AlHaz | just guessing |
| 17:13:25 | AlHaz | er, two banks of 8 instead of 4 banks of 4 |
| 17:14:41 | SpaceCoaster | yeah the SA1110 datasheet mentions that it can have two banks with DIFFERENT sizes, like it is a big feature, so maybe the SA1100 is limited |
| 17:14:44 | AlHaz | crucial says the sodimm is 8 meg * 32, I dunno what the chips on the board are . . . |
| 17:15:40 | AlHaz | I'm gonna go make a sandwitch |
| 17:15:49 | SpaceCoaster | I also have a rare sodimm from crucial which has 8k refresh or pages instead of 4k this doesn't work and I was hoping to test that as well |
| 17:15:55 | SpaceCoaster | me too |
| 17:28:10 | GPSFan | SpaceCoaster: has anyone read out the memory control registers while running Inferno with a 32M SODIMM? |
| 17:31:11 | ctr | oh my god |
| 17:31:18 | ctr | can the speaker now be used within linux? |
| 17:31:23 | ctr | ie with like speaker phone |
| 17:34:55 | mallum | AlHaz: if a devices firmware can boot an image over tftp, then it should be able to boot the same image over something else - eg disk ? |
| 17:34:55 | AlHaz | not quite yet |
| 17:35:00 | mallum | is back (gone 00:36:19) |
| 17:35:26 | AlHaz | mallum: theoretically, I guess, if the firmware can read from something else |
| 17:37:58 | mallum | AlHaz: I think the firmware can read ext2 |
| 17:38:42 | AlHaz | mallum: that'd be pretty convenient, does it have a block driver for ide or something? does it understand partitions? |
| 17:39:16 | mallum | AlHaz: I guess so |
| 17:39:56 | mallum | AlHaz: Im sure I remember reading somewhere it would read ext2 |
| 17:39:56 | AlHaz | should be pretty simple to do then |
| 17:40:03 | AlHaz | this the atx strongarm board? |
| 17:40:12 | mallum | yep |
| 17:40:35 | mallum | Its recognised the 256 meg of ram which is good |
| 17:40:51 | mallum | just didn't think installing would be this much of a pain |
| 17:42:43 | mallum | AlHaz: http://www.chaltech.com/linux.php - saying ext2 is supported but not partitioning enablement |
| 17:43:24 | AlHaz | mallum: Hmm |
| 17:43:25 | mallum | AlHaz: does that mean it'll be ok with one big partion ? |
| 17:43:47 | AlHaz | mallum: maybe don't partition the drive at all? idunno |
| 17:51:51 | mallum | AlHaz: new idea - I'll install netBSD ( that boots ok from CD ), copy the tftp image onto its root and then boot off that. |
| 17:58:15 | as_ | Evnin all. |
| 17:58:35 | collord | didn't know there was SMP ARM boards... does linux support this? |
| 17:59:03 | as_ | There have been SMP arm board for a long time. |
| 17:59:32 | as_ | One was called 'hydra' or somthing similar. Supported 16 CPUs. |
| 17:59:52 | collord | Risc OS? |
| 17:59:58 | as_ | Yes. |
| 18:00:16 | collord | exposes his ignorance once again |
| 18:00:20 | as_ | I dont know if it worked with ordinary programs though. |
| 18:00:35 | as_ | I suspect it needed specific programs to use the extra CPUs. |
| 18:01:05 | SpaceCoaster | GPSFan: don't know about memory control registers inferno or otherwise, do you any info? |
| 18:02:02 | as_ | The 'RiscPC' was dual CPU design intentionally one x86 and one ARM, I think the hydra plugged into that. |
| 19:06:28 | GoRK | argh i am just dumb |
| 19:07:05 | GoRK | i want to write one predetermined byte to a fd. everything i do segfaults |
| 19:08:48 | GoRK | wait i think i got it |
| 19:08:51 | GoRK | duh |
| 19:08:54 | Alehaz | ? |
| 19:09:16 | GoRK | write() takes a pointer. im stupid |
| 19:09:52 | GoRK | yep |
| 19:09:58 | GoRK | i am now keeping the dsp alive |
| 19:10:17 | SpaceCoaster | Did you see Bill Danielsons post? |
| 19:10:28 | GoRK | nope just got back from some lunch |
| 19:11:11 | SpaceCoaster | I tracked him down from the Wiki (and Google) and he scoped inferno it sends a b7 once per second |
| 19:11:18 | GoRK | SpaceCoaster: this watchdog thing only has to happen every 2-3 seconds it seems... userland will not be a problem. I send it sanity check commands and it likes those |
| 19:11:26 | GoRK | SpaceCoaster: that's exactly what i am doing |
| 19:11:40 | GoRK | # ./buzby a3 44 00 00 0f a7 06 22 1a a6 04 58 0e 58 03 58 07 58 02 58 02 58 06 58 01 58 01 |
| 19:11:40 | GoRK | 80 80 80 80 80 80 80 80 80 80 80 80 |
| 19:11:40 | GoRK | 80 db 06 f9 |
| 19:11:40 | GoRK | 80 db 06 f9 22 35 |
| 19:11:41 | GoRK | 80 db 06 f9' |
| 19:11:43 | SpaceCoaster | inspired or what? |
| 19:11:43 | GoRK | etc |
| 19:11:57 | GoRK | well.. i just figured that was the best thing to do |
| 19:12:13 | GoRK | since it's the only command that doesnt make the dsp do anything except tell you that it's woring or not |
| 19:12:34 | SpaceCoaster | We could just throw away the 80s |
| 19:12:48 | GoRK | they are important. they are command acks. |
| 19:12:54 | GoRK | 84 is a nak |
| 19:13:06 | SpaceCoaster | yeah but you can't really tell what they ack |
| 19:13:20 | GoRK | you can if you send the commands one at a time like inferno does |
| 19:13:35 | GoRK | they will ack and nak in sequence |
| 19:14:24 | SpaceCoaster | you would have to wait and timeout, but I bet you wouldn't do anything different for either an ack or a nak |
| 19:14:29 | SpaceCoaster | in most cases |
| 19:15:43 | GoRK | well still shouldnt throw it away |
| 19:17:01 | SpaceCoaster | yeah I agree but inferno does, I doubt if matching them up will give any positives |
| 19:17:17 | GoRK | and it looks like the dsp reset on device open is indeed the way to go |
| 19:17:32 | GoRK | the inferno kernel driver does? |
| 19:17:39 | GoRK | or userspace stuff? |
| 19:18:39 | SpaceCoaster | right, we have interrupts going and are doing a funky light weight fixed size queue, it will store N messages and wrap around deleting old ones. |
| 19:19:08 | GoRK | do you have some new code yet that works? |
| 19:19:16 | GoRK | i was going to start writing the daemon |
| 19:19:49 | SpaceCoaster | not yet, almost but need to take a break, polling etc. will work by Monday, maybe tomorrow |
| 19:20:13 | GoRK | cool well i can write a driver that just continuously tries to read in order to get something working |
| 19:20:36 | SpaceCoaster | Does a 64 read message queue sound good? |
| 19:20:47 | GoRK | no you'll need 650 or so |
| 19:20:51 | GoRK | let me look exactly |
| 19:21:21 | GoRK | speakerphone param+state data will all come down at once when you ask for them |
| 19:22:25 | SpaceCoaster | i didn't explain it well, the data for the message is malloced, 64 is the max number of unread messages before we start throwing old ones away. |
| 19:22:32 | GoRK | oh |
| 19:22:49 | GoRK | so you would be queuing that 650 byte message as one slot of 64? |
| 19:22:57 | GoRK | err actually it would be 2 slots |
| 19:23:00 | SpaceCoaster | very small messages are not malloced, they are stored in the pointer which would be used for the malloced data, sneaky eh |
| 19:23:17 | GoRK | hey, however you want to do it as long as they are spit out the same way :) |
| 19:23:46 | GoRK | i figure if you let 64 messages build up before you want to look at the responses, your code might as well not even work :) |
| 19:23:58 | SpaceCoaster | yup the 650 would use one of the slots |
| 19:24:06 | SpaceCoaster | your code or my code? |
| 19:24:26 | GoRK | whoever's code is dumb enough not to ever read from the buffer |
| 19:24:35 | GoRK | userland code |
| 19:26:02 | SpaceCoaster | k, each message uses exacly one slot, its only funky when we need to allocate space for long messages. The acks an naks will use 1 slot each, so 64 should last us a couple of minutes. |
| 19:27:07 | SpaceCoaster | The writes will not be queued, they are synchronous, when it returns it's gone. |
| 19:27:26 | GoRK | well we're going to have to be able to get that message from a keypress pretty fast so that we can process it and return |
| 19:28:07 | GoRK | hmm there are seperate events for key down and key up. that's useful |
| 19:29:03 | SpaceCoaster | Yeah that's what I thought, should be able to do neat things with the keypad. I'm glad that the designers didn't use that 10c chip to generate DTMF, this is more fun. |
| 19:29:10 | GoRK | the button up event is the same though |
| 19:29:14 | GoRK | for each key |
| 19:29:20 | SpaceCoaster | yes |
| 19:29:57 | GoRK | it wont read more than one key at a time |
| 19:30:14 | SpaceCoaster | thats good! |
| 19:30:24 | GoRK | i have listened to the stupid weather about 1 billion times testing this :) |
| 19:30:40 | GoRK | looking at hex dumps |
| 19:31:35 | SpaceCoaster | the sun is shining here |
| 19:32:17 | GoRK | it's partly cloudy here with a high of 80 degrees. winds south to southwest at 5-10mph |
| 19:32:44 | GoRK | and cars are on sale |
| 19:33:19 | SpaceCoaster | is that a guess? 80s bad,bad it's 58.5 here and that is just fine |
| 19:34:07 | SpaceCoaster | GoRK: which state are you in, UT? |
| 19:34:11 | GoRK | tx |
| 19:34:33 | SpaceCoaster | ah, near prplplague? |
| 19:34:47 | GoRK | where is he? im in amarillo.. in the panhandle |
| 19:35:12 | SpaceCoaster | austin, maybe |
| 19:35:45 | GoRK | hmm that's pretty far away |
| 19:35:52 | SpaceCoaster | no he is in tyler, his mom is in austin! |
| 19:36:13 | GPSFan | SpaceCoaster: sorry for the looong delay, if you read section 10.1 of the SA1100 datasheet, it talks about setting up the memory control registers etc. I've looked at where blob does this and it is in memsetup.S. My thinking is that if Inferno works with larger SODIMMS why not read out those registers and see what they are set to. Then there is the MMU setup, and sizing, but it shouldn't be too hard to figure out from reading a working set of registers. |
| 19:36:19 | SpaceCoaster | definitely a stroll from NH |
| 19:37:06 | GPSFan | SpaceCoaster: Inferno remote rdb can read that data. |
| 19:37:35 | SpaceCoaster | GPSFan: I tried to reinstall inferno, three times and failed, could you do it? I guess you TuxScreen is in bits! |
| 19:37:57 | GPSFan | SpaceCoaster: Bits & Pieces..... |
| 19:38:23 | GPSFan | SpaceCoaster: I' sure there are some Inverno's ldft ;>) |
| 19:38:49 | GPSFan | My typing is getting bad, time for a beer! |
| 19:39:19 | SpaceCoaster | GPSFan: When I reinstalled inferno I got a System V login on the serial port for a machine called Dante! |
| 19:40:45 | GoRK | SpaceCoaster: are you serious?!? thats crazy |
| 19:41:17 | GPSFan | SpaceCoaster: BTW, the schematics for the memory subsystem indicate that there are 4 banks of DRAM, 2 onboard, and 2 on the SODIMM. |
| 19:41:20 | SpaceCoaster | yes serious and crazy, so true. |
| 19:41:33 | SpaceCoaster | k |
| 19:43:33 | GPSFan | SpaceCoaster: addresses 0xA0000000-0xA000000C are the DRAM config and DRAM CAS waveform registers. |
| 19:45:25 | SpaceCoaster | GPSFan: are these on the SA1100? |
| 19:45:44 | GPSFan | SpaceCoaster: Yes. |
| 19:47:10 | SpaceCoaster | found the section, ok |
| 19:47:41 | GPSFan | Look at 10.2.1 and 10.3.1 |
| 19:48:33 | SpaceCoaster | that is definitely the right stuff! |
| 19:49:01 | GPSFan | Then look at memsetup.S in blob, that is where those reg's are setup. |
| 19:49:51 | SpaceCoaster | right, got it |
| 19:50:10 | GPSFan | blob just assumes 8M on board and 8M on SODIMM, following the 1M x 16 DRAM config described in 10.3.1 |
| 19:51:48 | GPSFan | The Tux ties A12 of the SODIMM to nCS1 which is non-standard, so the Linux memory management code needs to know this in order to toggle it when it needs to get at higher mem on the SODIMM, |
| 19:52:29 | SpaceCoaster | oh sortof manually paged memory? |
| 19:52:34 | GPSFan | Yep. |
| 19:52:44 | SpaceCoaster | why oh why oh why |
| 19:53:10 | GPSFan | I guess A12 must not participate in the refresh. |
| 19:54:08 | SpaceCoaster | I would let it "not participate" it either pulls it weight or goodbye, hardware engineers are too easy going! |
| 19:54:21 | SpaceCoaster | s/would/would not/ |
| 19:54:59 | SpaceCoaster | so this is probably not a simple blob issue at all? |
| 19:58:25 | GPSFan | I don't really know, if Inferno can work with >16M Linux should be able to as well. A12 is also on much larger DIMMS. the one I was looking at was the Crucial 64M, so A12 may not be importand for smaller memories. |
| 19:59:24 | GPSFan | SpaceCoaster: the other thing about the SODIMM socket is that there a lot of connections to pins labeled no-connect on the Crucial spec. |
| 20:02:14 | SpaceCoaster | GPSFan: thanks for the info, I will have a mooch around and see if I can understand some more, this has been enlightening, I've got it logged so I'll review it a bit later, thanks again, must enhance this driver for Lenolium and GoRk. |
| 20:03:48 | GoRK | heheeh |
| 20:04:12 | GPSFan | SpaceCoaster: ok great, I'm still working on schematics, I'v got all the GPIO pins located, and am trying to finish up the gate level logic on the CPU baord. |
| 20:04:18 | GoRK | i have no idea how far along lenolium is.... we are probably duplicating each others work at this point |
| 20:04:37 | GoRK | GPSFan: ucb1200 <-> mongoose <-> phone lines :)))))))))))) |
| 20:04:52 | GoRK | hehehehe |
| 20:04:59 | SpaceCoaster | miracles |
| 20:05:42 | GoRK | all i need to figure out is if and where the extra LED control pins are so i can connect a relay to them that switches the audio into the UCB1200 |
| 20:06:02 | GoRK | and where the ucb1200 audio in/outs are.. i couldnt locate the correct datasheet to show me that |
| 20:07:02 | SpaceCoaster | GoRK: isn't that in the UCB1200 product spec? |
| 20:07:15 | GoRK | yeah i just couldnt find it |
| 20:07:26 | GoRK | err i didnt look very hard either :) |
| 20:07:36 | SpaceCoaster | k |
| 20:08:39 | GoRK | i am thinking that the DSP does have the capability to record and play back to its onboard flash ram |
| 20:09:03 | GPSFan | GoRK: do you have the UCB1200 data sheet? |
| 20:10:40 | GPSFan | GoRK: try page 34. |
| 20:11:42 | GPSFan | GoRK: for the telcom, and page 8 for the microphone. |
| 20:13:35 | GoRK | blah |
| 20:13:37 | GoRK | anyway |
| 20:13:39 | GoRK | the ucb1200 data sheet is on the wiki somewhere |
| 20:13:41 | GoRK | hang on |
| 20:14:31 | GoRK | http://www-us2.semiconductors.philips.com/pip/ucb1200be |
| 20:54:25 | GoRK | haha i just called up my buddy and played happy tone and sad tone through the line for him |
| 20:54:39 | GoRK | it worked too :) |
| 21:14:30 | SpaceCoaster | hehe |
| 21:15:08 | SpaceCoaster | GoRK: do you have caller ID? |
| 21:20:41 | GoRK | SpaceCoaster: i wont be able to do that until i have some sort of daemon worked up |
| 21:20:57 | GoRK | SpaceCoaster: if you mean do i subscribe to caller id service, then yes. i do and i have callerid on call waiting |
| 21:21:11 | GoRK | SpaceCoaster: right now i am typing out a header file |
| 21:21:27 | GoRK | somehow i didnt think we could just use shandsp.h |
| 21:21:30 | GoRK | :) |
| 21:22:08 | GoRK | oh and i did manage to get one led in all 4 states. (flashing and blinking timers are seperate and do work) -- the inferno code was broken. |
| 21:22:26 | GoRK | ./buzby A3 04 00 0B D0 <-- one led in each of 4 states, 2 flashing speeds. not recommended for epileptics |
| 21:25:01 | SpaceCoaster | nice |
| 21:28:48 | GoRK | that is with my modified buzby though... that sends the sanity check every second after the commandline |
| 21:28:53 | GoRK | dunno if it makes a difference |
| 21:30:03 | SpaceCoaster | GoRK: should I produce the next driver patch base on the previous or on the linux...tux1 kernel? |
| 21:30:54 | GoRK | well... i'd apply it on top of linux+rmk+tux+touchscreen |
| 21:31:19 | GoRK | since doesnt it have to remove the lines from the touchscreen patch where it resets the dsp a bunch of times? |
| 21:32:03 | GoRK | ideally it would just apply on top of linux+rmk+tux... or be part of the tux patch |
| 21:32:06 | SpaceCoaster | not essential but a good idea, ok so you will have to back out the wheaties0 patch |
| 21:32:22 | GoRK | no worries |
| 21:33:10 | GoRK | i was able to make my LPP patch apply at any stage... linux, linux+rmk, linux+rmk+tux, but it was half just lucky |
| 21:40:38 | GoRK | well i gotta go for a while |
| 21:40:45 | GoRK | talk to everyone later tonight |
| 21:40:51 | SpaceCoaster | see ya later |
| 21:40:56 | GoRK | bye |
| 21:42:07 | GoRK | your kernel patch should probably include this header file i'm writing |
| 21:42:14 | GoRK | but we'll worry about that later |
| 21:42:22 | SpaceCoaster | k |
| 21:42:43 | SpaceCoaster | I have the kernel in cvs so patches are easy to produce |
| 22:03:06 | wej | == wedge |
| 22:04:48 | wej | anyone alive? |
| 22:09:00 | scanline | is away: food |
| 22:10:13 | SpaceCoaster | hi wej |
| 22:11:24 | wej | how goes it, SC? |
| 22:11:40 | SpaceCoaster | good, had some DSP progress, did you hear? |
| 22:12:03 | wej | haven't heard. |
| 22:12:05 | wej | how's it goin? |
| 22:13:01 | SpaceCoaster | bleepin', blinkin', dialin' just swearin' n' cusin' to do and were done. |
| 22:15:16 | wej | very nice. |
| 22:15:33 | wej | full dup? |
| 22:15:48 | SpaceCoaster | GoRK: is doing investigating the commands I'm doing the driver. |
| 22:16:01 | wej | right on. |
| 22:16:03 | wej | damn you guys rule. |
| 22:16:26 | wej | i'm gonna bite the bullet today and get a supported 802.11 card. |
| 22:16:46 | wej | RedBaron did a microwindows mod to enable screen dimming |
| 22:16:47 | SpaceCoaster | full dup?, it isn't transfering audio, just controls and tones. |
| 22:16:48 | wej | did you see it? |
| 22:17:00 | wej | ah . gotcha. |
| 22:17:09 | SpaceCoaster | screen dimming wotzat? |
| 22:17:35 | SpaceCoaster | aahh the keys I remember |
| 22:17:48 | wej | yeah. |
| 22:18:00 | wej | i've known the guy for a while. |
| 22:18:09 | wej | he's gonna be on that robot battle competition show soon. |
| 22:18:15 | wej | he built this pretty sweet bot. |
| 22:19:01 | SpaceCoaster | he has some interesting ideas for a mobile TuxScreen |
| 22:20:14 | wej | yeah. he got it working off of batteries. |
| 22:20:45 | SpaceCoaster | I might buy one and do the same, needs the wireless though, which card are you buying? |
| 22:21:00 | wej | probably a CF card |
| 22:21:13 | wej | the linksys card is not working. |
| 22:21:16 | wej | still crashing the tuxscreen. |
| 22:21:19 | wej | driving me insane. |
| 22:21:23 | wej | tuxscreen is ... so... close. |
| 22:21:46 | SpaceCoaster | sorry I couldn't help, just sympathise |
| 22:21:52 | wej | no problemo amigo. |
| 22:21:55 | wej | i appreciate you looking into it. |
| 22:22:03 | wej | btw.. drink something resembling a cleaning product for me tonight. |
| 22:22:06 | wej | it's my birthday n'shit. |
| 22:23:28 | SpaceCoaster | congrats, cleaning products are the only thing in the house to drink, so here goes the windex, glug, glug, glug |
| 22:23:57 | SpaceCoaster | hey, it cleaned my contact lenses from the inside! |
| 22:24:27 | wej | hehehe. |
| 22:24:34 | wej | easy on that 409, man. |
| 22:24:36 | wej | gives me serious gas. |
| 22:25:22 | SpaceCoaster | my wife found two drops, literally of single malt whisky, cheers, happy birthday! |
| 22:26:39 | wej | just ordered another tuxscreen. |
| 22:26:42 | wej | heeheh. |
| 22:26:44 | wej | thanks, man :) |
| 22:26:48 | wej | and thank yer wife too :) |
| 22:26:52 | wej | what kinda single malt whiskey? |
| 22:27:43 | SpaceCoaster | Lagavulin from Isley in Scotland |
| 22:28:02 | wej | my favorite. |
| 22:28:05 | wej | they make Oban as well. |
| 22:28:08 | wej | that's a really nice single malt. |
| 22:29:16 | SpaceCoaster | yes, very peaty and smokey, sometimes too strong, have to be in the right mood, we also bought some mcclelland's which was cheap but very good, bit lighter |
| 22:30:05 | SpaceCoaster | we went to Oban a few times, nice little town, the west coast of Scotland is heaven |
| 22:32:25 | wej | very nice. |
| 22:32:30 | wej | i didn't realize it was a town. |
| 22:32:36 | wej | <- american |
| 22:32:50 | SpaceCoaster | small fishing village |
| 22:32:55 | wej | sounds cool. |
| 22:32:59 | wej | lagavulin is a town as well, or no? |
| 22:33:28 | wej | i'v egot this cool opportunity to move to London for 6 months, and then on to Lyon France. |
| 22:33:37 | wej | this company i started out is getting some serious traction. |
| 22:33:39 | wej | totally unexpected. |
| 22:34:52 | SpaceCoaster | great, London is a big city, fun but hectic, have you been to Europe at all? |
| 22:35:03 | wej | yeah. |
| 22:35:13 | wej | i've been to london a couple of times and i worked in south france for about a month |
| 22:35:52 | SpaceCoaster | wej != regular american |
| 22:36:11 | scanline | is back (gone 00:27:11) |
| 22:36:14 | SpaceCoaster | what does your company do? |
| 22:36:24 | wej | heh. thanks, man. |
| 22:36:36 | wej | we're working on a product for network architecture.. |
| 22:36:38 | wej | all ASP based.. |
| 22:36:42 | wej | written in Java/Swing |
| 22:37:02 | wej | kinda a cross between their windows Cisco ConfigMaker and the NetSys network modeler |
| 22:37:18 | SpaceCoaster | ASP, Java/SWing do they go together? |
| 22:37:33 | wej | yeah.... |
| 22:37:38 | wej | the client is very lite |
| 22:38:07 | wej | will ultimately be web based if possible.... will need the swing classes in the browser tho... |
| 22:38:29 | wej | the company i worked for that had an office in south france made an ASP based iPBX |
| 22:38:31 | wej | all written in java.. |
| 22:38:36 | wej | except for a protocol translator my group did. |
| 22:38:40 | wej | did h.323->SIP |
| 22:38:48 | SpaceCoaster | cool, the jdk1.3 stuff and jnlp should take care of the browser side |
| 22:39:00 | wej | hopefully, yeah. |
| 22:39:08 | wej | 1.3 is pretty fast, too. |
| 22:39:29 | SpaceCoaster | yeah, thats what we will be running on the TuxScreen |
| 22:39:35 | wej | right on. |
| 22:40:17 | SpaceCoaster | All this driver stuff is just clearing the way for a Java/Swing app. It's fun too. |
| 22:40:29 | wej | man... that'll be sweet. |
| 22:41:03 | SpaceCoaster | It needs the memory though, so I need to look at the problem with 32mb cards and linux. |
| 22:41:29 | wej | think that's coming soon? |
| 22:42:30 | SpaceCoaster | maybe, it's next on the list, I think I will be done with the DSP in a couple of days, enough to let people test/develop and discover |
| 22:43:25 | SpaceCoaster | have you used jsp/servlets much? |
| 22:43:38 | wej | a little, yeah. |
| 22:44:05 | SpaceCoaster | did you like it? |
| 22:44:07 | wej | they can be pretty sweet. it's need for network stuff.. tying into DBs on the back end n'all. |
| 22:44:12 | wej | i'm more fond of doing that stuff in perl tho, really :) |
| 22:44:19 | wej | but in an all java environment, i was forced to play with it some. |
| 22:44:37 | wej | applets, unless signed, don't allow you to open socket conections to other hosts... do things like back end database routines... |
| 22:44:47 | wej | servlets let you bypass that stuff... |
| 22:44:52 | wej | and the networking code is so sweet in java... |
| 22:45:40 | SpaceCoaster | true, I tried java a few times early on but it wasn't good enough, now it really is sweet |
| 22:46:40 | SpaceCoaster | somethings, imageio for one, keep changing but they get better each time, so can't really complain. |
| 22:51:44 | wej | just always a pain in the butt to go back and rewrite a zillion deprecated methods. |
| 22:54:26 | SpaceCoaster | true, got any birthday plans? |
| 22:55:07 | wej | my wife is doing a little Tapas thing tonight. |
| 22:55:14 | wej | i was hoping she had said 'topless' |
| 22:55:18 | wej | ah well. |
| 22:55:33 | SpaceCoaster | lol |
| 22:58:04 | SpaceCoaster | wej: we seem to have three groups here, texans, utahians and brits, which are you? |
| 23:04:56 | SpaceCoaster | wej: gonna go eat, happy birthday, good luck with the tapa, see ya |
| 23:12:14 | wej | funny how such a wonderful thing as the tuxscreen can put you in touch with the coolest people. |
| 23:12:30 | wej | feels the need to hug his tuxscreen. |
