fluid-work IRC Logs-2010-07-21

Owned by Fluid IRC Log Bot
Last updated: Jul 21, 2010Version comment
9 min read

[03:36:53 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined #fluid-work
[05:20:29 CDT(-0500)] * kasper (~kasper@0x57350a51.odnxx12.dynamic.dsl.tele.dk) has joined #fluid-work
[06:09:18 CDT(-0500)] * zafar (~zafar@2001:638:208:4807:a6ba:dbff:fe02:2266) has joined #fluid-work
[07:24:48 CDT(-0500)] * michelled (~michelled@142.150.154.141) has joined #fluid-work
[07:40:09 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[07:51:03 CDT(-0500)] * anastasiac (~team@142.150.154.193) has joined #fluid-work
[08:00:39 CDT(-0500)] * jessm (~Jess@c-68-40-84-116.hsd1.mi.comcast.net) has joined #fluid-work
[08:27:14 CDT(-0500)] * jhung (~Jon@H233.C205.cci.switchworks.net) has joined #fluid-work
[08:29:10 CDT(-0500)] * colinclark (~colin@bas2-toronto09-1176131242.dsl.bell.ca) has joined #fluid-work
[08:47:05 CDT(-0500)] * clown (~clown@142.150.154.103) has joined #fluid-work
[08:49:25 CDT(-0500)] * sgithens (~sgithens@149-166-10-223.dhcp-in.iupui.edu) has joined #fluid-work
[09:28:48 CDT(-0500)] * mackrauss (~Armin@142.150.154.103) has joined #fluid-work
[09:34:30 CDT(-0500)] * mackrauss (~Armin@142.150.154.103) has left #fluid-work
[09:57:51 CDT(-0500)] <jhung> justin_o, colinclark: are the install scripts ready to be used for testing. i.e. if I used it now, will it grab the right versions we should be using for testing?
[09:58:07 CDT(-0500)] <Justin_o> jhung: should be, james and I have started testing already
[09:58:07 CDT(-0500)] <colinclark> jhung: Yep
[09:58:29 CDT(-0500)] * bsparks (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined #fluid-work
[09:58:40 CDT(-0500)] <Justin_o> jhung: what was the resolution of the images that you uploaded that are used for the --no-cameras capture
[09:59:09 CDT(-0500)] <colinclark> In case it helps, I have a fresh VMWare image with everything set up for testing
[09:59:16 CDT(-0500)] <colinclark> It's 1.5 Gb, so I'm not quite sure where to put it
[09:59:19 CDT(-0500)] * alison-b (~chatzilla@dsl-173-206-255-172.tor.primus.ca) has joined #fluid-work
[09:59:27 CDT(-0500)] <Justin_o> jhung, jameswy, colinclark: Also i noticed that we don't really have much in the way of focus styling. which makes using the keyboard rather difficult
[09:59:31 CDT(-0500)] <colinclark> But I was thinking that perhaps we could redistribute it as well as an option for using Deacpod
[10:00:07 CDT(-0500)] <jhung> justin_o: yes. Focus styling is missing. Rotations are especially hard.
[10:00:14 CDT(-0500)] <Justin_o> colinclark: that would be good... i'll check out google code to see if we can upload it there, not as part of the source code though
[10:00:20 CDT(-0500)] <Justin_o> jhung: yep...
[10:00:21 CDT(-0500)] <colinclark> Justin_o: I checked
[10:00:25 CDT(-0500)] <colinclark> max 200 Mb file size
[10:00:34 CDT(-0500)] <Justin_o> the question is that a blocker for this release or can we fix it afterwards
[10:00:34 CDT(-0500)] <Justin_o> {color}
[10:00:34 CDT(-0500)] <jhung> justin_o: resolution is 1514x2375
[10:00:36 CDT(-0500)] <colinclark> We probably have space on fluidproject.org
[10:00:43 CDT(-0500)] <colinclark> but it's a lot to be distributing
[10:01:09 CDT(-0500)] <Justin_o> colinclark: yep.. we should probably ask jamon what he thinks about that
[10:01:33 CDT(-0500)] <Justin_o> jhung: i'm noticiing that the stitching on those supplied images have white space in between them.
[10:02:15 CDT(-0500)] <jhung> justin_o: Not sure if the lack of focus is a blocker... but I can be pursuaded.
[10:02:33 CDT(-0500)] <colinclark> maybe, in the end, we should wait to distribute it as a VM until 0.5 when it will actually work
[10:02:58 CDT(-0500)] <colinclark> jhung: Justin_o and I have been sketching out a plan for a small 0.5 release immediately following this one
[10:03:01 CDT(-0500)] <jhung> Justin_o: some of the images have white space around the image. So stitching is doing fine.
[10:03:19 CDT(-0500)] <colinclark> Which will fill in some of those "blockers" you sketched out for me in terms of L/R capture
[10:03:24 CDT(-0500)] <colinclark> 1. Deskewing
[10:03:27 CDT(-0500)] <colinclark> 2. Cropping
[10:03:32 CDT(-0500)] <colinclark> 3. PDF export timeout
[10:03:53 CDT(-0500)] <colinclark> Giving you and jameswy a chance to absorb some of zafar's new stuff from the design and functionality perspective
[10:05:06 CDT(-0500)] <jhung> Colinclark: Cool. Any idea when 0.5 would be? If possible, we should integrate type 2 PDF output if we can figure out ocropus + genpdf issues with OCR.
[10:05:20 CDT(-0500)] <colinclark> that would be quite interesting
[10:05:26 CDT(-0500)] <Justin_o> jhung: thanks for the explanation about the stitching
[10:05:43 CDT(-0500)] <colinclark> would that involve much in the way of dev work for us, jhung, or would it mostly be interfacing with the IUPR team to get all the settings working, etc?
[10:06:09 CDT(-0500)] <Justin_o> colinclark: i think you are right that it might be better to wait till 0.5 to do the vmware image
[10:06:17 CDT(-0500)] <colinclark> Justin_o: k
[10:06:29 CDT(-0500)] <zafar> colin any firm date on 0.5 release?
[10:06:32 CDT(-0500)] <colinclark> In the meantime, I'll hold on to this so we have a clean image in case people need to test or use it
[10:06:45 CDT(-0500)] <colinclark> zafar: Well, we're just talking a bit about that now
[10:07:21 CDT(-0500)] <colinclark> We wwere saying that it might be good do a "mini release" immediately after 0.4 to get independent L/R capture working end to end with OCR
[10:07:30 CDT(-0500)] <colinclark> That would mostly keep me and Justin_o busy for a few weeks
[10:07:58 CDT(-0500)] <colinclark> While you, jhung, and jameswy simultaneously start work on 0.6, which would include stereo functionality
[10:08:10 CDT(-0500)] <colinclark> zafar: I'm assuming that would include both stereo calibration and dewarping?
[10:08:40 CDT(-0500)] <colinclark> In terms of timelines of the top of my head, it might be early Sept. for 0.5 and late Sept/early Oct for 0.6, but I don't know yet
[10:08:47 CDT(-0500)] <colinclark> We still have to do some estimating
[10:09:10 CDT(-0500)] <zafar> at the moment there is no dewarping, it's still in research phase, but stereo calib is there, just wondering will a calib gui wireframe be ready till "mini release"?
[10:09:45 CDT(-0500)] <jhung> zafar, jameswy: we'll need to come up with a new design for stereo calibration. We'll start planning that once this release is out.
[10:10:36 CDT(-0500)] <jhung> zafar: a wireframe has not been done yet for stereo, but that will be a priority once this release is done.
[10:11:46 CDT(-0500)] <jhung> zafar: we'll get together and talk about what the UI will do, and the kind of feedback / interactions the system has with the user.
[10:12:28 CDT(-0500)] <colinclark> Yep, that's right. So we'll start designing for stereo as soon as this release is out the door.
[10:12:41 CDT(-0500)] <zafar> ok sounds good
[10:15:06 CDT(-0500)] <Justin_o> colinclark, jhung: so for the focus styling issue do you guys want to tackle that in the 0.5 release as well
[10:17:03 CDT(-0500)] <colinclark> The issue is that we aren't styling keyboard focus very prominently?
[10:17:05 CDT(-0500)] <colinclark> Or at all?
[10:19:44 CDT(-0500)] <jhung> colinclark: none at all.
[10:19:52 CDT(-0500)] <colinclark> ok
[10:19:58 CDT(-0500)] <colinclark> sounds like one that's high on our list to fix
[10:20:06 CDT(-0500)] <colinclark> What would it take to fix it now, during code freeze?
[10:20:13 CDT(-0500)] <jhung> colinclark: it was actually removed intentionally because it was drawing large boxes for offscreen text too.
[10:20:26 CDT(-0500)] <colinclark> And, the usual question, what impact would fixing it have on our QA cycle?
[10:20:31 CDT(-0500)] <colinclark> jhung: can you elaborate?
[10:21:25 CDT(-0500)] <jhung> colinclark: the rotation buttons are links with background images of buttons. The link text is positioned offscreen so just the rotation image is viewable.
[10:21:47 CDT(-0500)] <jhung> But when you focus on the element, screen readers should be able to pick up the anchor text and read it off.
[10:22:05 CDT(-0500)] <colinclark> yeah
[10:22:10 CDT(-0500)] <colinclark> That's one we should fix
[10:22:12 CDT(-0500)] <jhung> but because the anchor text is off screen, the default focus style created this long horizontal boundind box.
[10:22:14 CDT(-0500)] <colinclark> But assume it'll take some work?
[10:22:31 CDT(-0500)] <Justin_o> jhung: for those we could have a different image when focused
[10:22:34 CDT(-0500)] <colinclark> jameswy: Rumor has it you found a blocker?
[10:22:48 CDT(-0500)] <jhung> colinclark, justin_o: exactly. We would use a different image.
[10:22:54 CDT(-0500)] <colinclark> k
[10:22:57 CDT(-0500)] <jhung> but keep the outline to zero.
[10:23:01 CDT(-0500)] <colinclark> jhung: Any guesses on a time to fix?
[10:23:19 CDT(-0500)] <jhung> ask jameswy. he has the assets to create the new icons.
[10:23:30 CDT(-0500)] <jhung> Styling should be quick.
[10:24:08 CDT(-0500)] <jameswy> colinclark: Rumor has it right. Export produces empty PDFs.
[10:24:17 CDT(-0500)] <colinclark> ?!?
[10:24:20 CDT(-0500)] <colinclark> That one must be super new
[10:24:41 CDT(-0500)] <jameswy> colinclark: We think it might be because we're using the older genpdf now. Different parameters?
[10:24:45 CDT(-0500)] <Justin_o> colinclark: yes... probably because of the roll back in the version of genpdf
[10:24:47 CDT(-0500)] <colinclark> interesting
[10:25:12 CDT(-0500)] <colinclark> ok, super interesting
[10:25:30 CDT(-0500)] <colinclark> looks like our advice from Michael was for a newer version of genpdf
[10:26:21 CDT(-0500)] <jameswy> jhung, colinclark: re: fixing focus styling: less than an hour for additional icons
[10:26:45 CDT(-0500)] <colinclark> ah, that sounds pretty good
[10:26:58 CDT(-0500)] <jhung> jameswy: make it so.
[10:26:59 CDT(-0500)] <colinclark> if we have to reopen the code anyway
[10:27:09 CDT(-0500)] <colinclark> hang on one sec, jhung
[10:27:11 CDT(-0500)] <jhung> (smile)
[10:27:13 CDT(-0500)] <colinclark> let's figure out the second question (smile)
[10:27:25 CDT(-0500)] <colinclark> Justin_o: What's the impact of the styling fix on our QA cycle?
[10:28:56 CDT(-0500)] <Justin_o> colinclark: hmm... i'll guess a quick run through everything keyboard only should be okay
[10:28:56 CDT(-0500)] <colinclark> jameswy: Can you toss up a stack trace or any other information on your blocker JIRA and send it along to the channel?
[10:28:59 CDT(-0500)] <colinclark> I'll take a look
[10:29:02 CDT(-0500)] <colinclark> Justin_o: Cool
[10:29:13 CDT(-0500)] <colinclark> jhung: Ok, now go for it
[10:30:28 CDT(-0500)] <Justin_o> colinclark, jhung, jameswy: making the styling issue a blocker for the release
[10:30:40 CDT(-0500)] <jhung> ok
[10:30:45 CDT(-0500)] <jameswy> colinclark: Yep. I'll put whatever information I have and put it into the JIRA (channel is a bit inconvenient right now since Decapod and my computer are on different desks)
[10:31:00 CDT(-0500)] <colinclark> jameswy: yeah, that's a drag
[10:47:55 CDT(-0500)] * elicochran (~elicochra@dhcp-169-229-212-70.LIPS.Berkeley.EDU) has joined #fluid-work
[10:48:28 CDT(-0500)] * jhung1 (~decapod@H233.C205.cci.switchworks.net) has joined #fluid-work
[10:48:48 CDT(-0500)] <colinclark> Justin_o: Ok, so I guess step #1 is to try to reproduce this issue
[10:50:43 CDT(-0500)] <jhung1> colinclark: I am using 150:c099afaea787 of Decapod server (tip)
[10:51:25 CDT(-0500)] <jhung1> colinclark: version 37:1434f02f66b2 of genpdf.
[10:51:25 CDT(-0500)] <Justin_o> colinclark, jhung: I was able to reproduce the issue in no camera mode
[10:51:37 CDT(-0500)] <jhung1> what exactly is the issue?
[10:53:21 CDT(-0500)] <Justin_o> jhung1: well the export is empty
[10:53:47 CDT(-0500)] <jhung1> Hmmm. That's odd. My output is fine. Let me try again...
[10:54:06 CDT(-0500)] <Justin_o> jhung1: did you run the new install scripts
[10:54:21 CDT(-0500)] <jhung1> justin_o yes.
[10:54:29 CDT(-0500)] <Justin_o> also i think your version of genpdf is wrong since it should have the tag
[10:54:34 CDT(-0500)] <Justin_o> if you do hg identify
[10:56:07 CDT(-0500)] <jhung1> justin_o: yeah, it's tell me it's tip, and not a tag.
[10:56:39 CDT(-0500)] <jameswy> colinclark, jhung, Justin_o: I've commented on the export blocker--added the console output
[10:56:50 CDT(-0500)] <colinclark> jameswy: What's the ticket number?
[10:56:51 CDT(-0500)] <jameswy> colinclark, Justin_o, jhung: DECA-153
[10:56:57 CDT(-0500)] <Justin_o> okay.... the decapod-decapod.sh install script should have done an update to the tag
[10:57:43 CDT(-0500)] <colinclark> jameswy, Justin_o, jhung: It looks pretty likely that this is related to Michael's advice to not use the -b flag
[10:57:54 CDT(-0500)] <colinclark> Clearly the API for genpdf changed between 0.3 and tip
[10:58:04 CDT(-0500)] <colinclark> So hopefully it will work if we use the old style arguments
[10:58:14 CDT(-0500)] <colinclark> But the one catch is that we also removed the TIFF conversion process
[10:58:22 CDT(-0500)] <colinclark> So we used to convert everything to TIFFs and make a multipage TIFF
[10:58:35 CDT(-0500)] <colinclark> Michael and Thomas said this wasn't necessary
[10:58:42 CDT(-0500)] <colinclark> Hopefully it was equally unnecessary back in the 0.3 era
[10:58:56 CDT(-0500)] <colinclark> I think we can probably fix this quickly
[10:59:01 CDT(-0500)] <jhung1> yes. we can feed it a string of pngs or jpegs
[10:59:18 CDT(-0500)] <colinclark> Do you know if we could we back in 0.3?
[10:59:58 CDT(-0500)] <jhung1> colinclark: It was added around the time of that release. Not sure if it made it into the release proper.
[11:00:07 CDT(-0500)] <colinclark> let's cross our fingers
[11:05:49 CDT(-0500)] <Justin_o> colinclark, jhung, michelled, anastasiac, jameswy: I'm about to send off the e-mail about our new e-mail addresses... did i get your new e-mail addresses correct
[11:06:04 CDT(-0500)] <michelled> yep
[11:06:10 CDT(-0500)] <anastasiac> yep
[11:06:45 CDT(-0500)] <jhung> justin_o: when did you send the email? I have it forwarding to my gmail, so it takes a little while.
[11:08:17 CDT(-0500)] <Justin_o> oh i sent one out to you guys a few weeks ago
[11:08:21 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined #fluid-work
[11:10:10 CDT(-0500)] <jhung> justin_o: what was the date of the email? I can't seem to find it.
[11:11:33 CDT(-0500)] <Justin_o> jhung: we're heading out to pho hung now for lunch
[11:11:38 CDT(-0500)] <Justin_o> jhung at ocad.ca
[11:11:48 CDT(-0500)] <Justin_o> that is what i had down for your e-mail
[11:11:54 CDT(-0500)] <Justin_o> talk to you in a bit
[11:48:53 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined #fluid-work
[11:54:08 CDT(-0500)] * bsparks1 (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined #fluid-work
[12:06:40 CDT(-0500)] <jhung1> jameswy: ping
[12:38:37 CDT(-0500)] * colinclark (~colin@142.150.154.103) has joined #fluid-work
[12:39:06 CDT(-0500)] <jhung1> ping colinclark
[12:39:44 CDT(-0500)] * michelled (~michelled@142.150.154.141) has joined #fluid-work
[12:40:46 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[12:40:55 CDT(-0500)] <colinclark> jhung1: what's up?
[12:41:15 CDT(-0500)] * yura_ (~yura@142.150.154.163) has joined #fluid-work
[12:42:10 CDT(-0500)] <jhung1> colinclark, i've rebuilt using the script. I am now getting version 36 as the genpdf script (not version 37). I run through capture and export, and everything is fine. The PDF has content in it.
[12:42:35 CDT(-0500)] <colinclark> Wait, I'm confused
[12:42:39 CDT(-0500)] <Justin_o> jhung1: is version 36 tagged at all?
[12:42:49 CDT(-0500)] <colinclark> You're saying you still can't reproduce jameswy's issue?
[12:43:16 CDT(-0500)] <jhung1> colinclark, justin_o: I'm still trying to confirm I have the right configuration here.
[12:43:43 CDT(-0500)] <jhung1> With the scripts by themselves in an empty directory tree, I run ./decapod-decapod.sh
[12:43:51 CDT(-0500)] <jhung1> That pulls down genpdf 36
[12:44:40 CDT(-0500)] <jameswy> jhung1: did you run a "remove" then "install", or a "clean" on the script?
[12:46:03 CDT(-0500)] <Justin_o> jhung1: the expect changeset for genpdf is 7a863b498d2e which is tagged as decapod-0.3/decapod-0.4
[12:46:37 CDT(-0500)] <Justin_o> which in human readable form is probably 13
[14:26:56 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[14:58:22 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[15:00:27 CDT(-0500)] * bsparks (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined #fluid-work
[15:11:21 CDT(-0500)] * colinclark_ (~colin@142.150.154.148) has joined #fluid-work
[15:12:39 CDT(-0500)] * anastasiac (~team@142.150.154.193) has left #fluid-work
[15:38:20 CDT(-0500)] * ihq (~ihq@212.101.229.29) has joined #fluid-work
[15:39:14 CDT(-0500)] <ihq> Hey everyone (good evening or good morning depending on where you are) can i ask a question on project Decapod here?
[15:39:42 CDT(-0500)] <ihq> or am I in the wrong channel?
[17:01:35 CDT(-0500)] * elicochran (~elicochra@dhcp-128-32-160-122.LIPS.Berkeley.EDU) has joined #fluid-work
[18:45:36 CDT(-0500)] * fathermojo (~elicochra@dhcp-169-229-212-61.LIPS.Berkeley.EDU) has joined #fluid-work
[20:26:49 CDT(-0500)] * jhung1 (~decapod@H233.C205.cci.switchworks.net) has left #fluid-work
[20:46:21 CDT(-0500)] * michelled (~michelled@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has joined #fluid-work
[20:46:58 CDT(-0500)] * jhung (~Jon@H233.C205.cci.switchworks.net) has joined #fluid-work
[20:50:22 CDT(-0500)] * jhung (~Jon@H233.C205.cci.switchworks.net) has left #fluid-work