Tuesday, 2014-06-10

*** tpb has joined #timvideos00:00
*** mparuszewski has quit IRC00:21
*** CarlFK has quit IRC03:47
*** CarlFK has joined #timvideos04:32
*** ChanServ sets mode: +v CarlFK04:32
*** mparuszewski has joined #timvideos05:21
*** slomo has joined #timvideos06:18
*** CarlFK has quit IRC06:40
*** rohitksingh has joined #timvideos06:55
mithroayush3504 / rohitksingh: ping?07:06
ayush3504mithro: pong07:06
mithroayush3504 / rohitksingh: meeting?07:06
*** mparuszewski has quit IRC07:31
*** sarwarc has joined #timvideos07:48
mithroayush3504 I'm back now08:22
ayush3504mithro: welcome back08:23
ayush3504i made the diagram in the same file for the wider version of the board https://docs.google.com/drawings/d/1_R3jtlCrZd9fILbPU8QGs2s61_4jhVgljF7P4RS27_U/edit08:24
tpbTitle: Serial expansion board sketch - Google Zeichnungen (at docs.google.com)08:24
mithroayush3504: yeah, I think that is a more reasonable layout08:24
ayush3504the width may or may not match the atlys board. it will depend on the minimum clearance between the db9 boards.08:25
ayush3504err, socket08:26
mithroayush3504: I think we should aim to match the Atlys board size. Otherwise we might want a different layout08:26
ayush3504ok. let's see. Any good suggestions to determine the clearance?08:27
ayush3504do you think there might be some standards for the clearance around serial port?08:28
mithroayush3504: nope! get out the ruler and measure?08:28
mithroayush3504: do a Google search?08:29
mithroayush3504: so your proposal for the daughter board would look like what I just added?08:29
ayush3504looking at the atlys board i don't think we can have more than 3 of those DB9 connectors based on the width of cables i have seen08:30
mithroayush3504: this has the problem that you need a "filler" rs232 daughter board if you want all the external connectors to align08:30
mithroayush3504: can you go vertical?08:30
ayush3504you mean like the left one, but with connectors on one side?08:31
mithroNo I mean stacked serial ports, http://www.alliedelec.com/search/productdetail.aspx?SKU=70115902 <-- usb version08:32
tpbTitle: Amphenol Commercial - UE27-AS54-13H - Data/Ethernet/Telecom Connectors - Connectors - Allied Electronics (at www.alliedelec.com)08:32
ayush3504do we have ports like that? I haven't seen them08:33
ayush3504yes I found them08:33
mithroI'm pretty sure they exist, but if they are cost effective is another matter08:34
ayush3504yes cost could be a matter, because the normal DB9s are common. Let me check.08:35
ayush3504mithro: maybe this one : http://in.mouser.com/ProductDetail/Amphenol-Commercial-Products/L17H2220131/?qs=sGAEpiMZZMtRCyOOFAXlCMxJPZK9qnVikjeCGXqo9wo%3d08:38
*** slomo has quit IRC08:38
tpbTitle: L17H2220131 Amphenol Commercial Products | Mouser (at in.mouser.com)08:38
ayush3504mithro: that's a female one, i better be careful buying these :/08:41
mithroayush3504: yes, you'll need to be careful buying anything :P08:43
mithroayush3504: so see the big red arrows I added to the diagram? This is the issue I was trying to explain regarding having the rs232 as daughter board08:44
ayush3504maybe we can keep the box till the db9s, and leave the rest to the user to customize.08:48
*** slomo has joined #timvideos08:51
*** slomo has joined #timvideos08:51
ayush3504mithro: the dual db9 connector goes around USD 6.6 http://www.digikey.com/product-search/en?mpart=178-009-613R571&vendor=3708:56
tpbTitle: 178-009-613R571 Norcomp Inc. | 1109MME-ND | DigiKey (at www.digikey.com)08:56
*** slomo has quit IRC08:57
mithroayush3504: how does that compare to non-dual DB9?08:58
mithroayush3504: and do you see/understand the problem with the current daughter board layout?08:58
ayush3504mithro: yes i got that. if we can limit the box till db9 then it may be solved.09:01
ayush3504mithro: single DB9 connector from a reputable mfr starts around USD 2 at digikey -  http://www.digikey.com/product-detail/en/1734354-1/A35107-ND/127982009:02
tpbTitle: 1734354-1 TE Connectivity | A35107-ND | DigiKey (at www.digikey.com)09:02
ayush3504mithro: so it's $6.6 vs $409:03
ayush3504mithro: $19.8 if you go with the stacked ones and $12 if you go with the regular ones09:05
mithroayush3504: put that in a spreadsheet :P09:06
ayush3504ha, okay!09:06
mithroayush3504: ---> do you see/understand the problem with the current daughter board layout?09:09
mithroayush3504: I added my proposal at the bottom of that diagram09:09
*** slomo has joined #timvideos09:09
*** slomo has joined #timvideos09:09
ayush3504mithro: is the bottom most drawing a detailed representation of the middle one?09:11
*** slomo has quit IRC09:14
ayush3504mithro: this looks good. Are you planning to have VHDCI at right side as shown in middle drawing or is it going to be at the opposite side as in the top-right drawing09:15
mithrothe bottom is more looking at the daughter board layout it makes not decisions about the direction of the layout09:15
mithroayush3504: I don't know which one is best with the ports facing the front or ports facing the side09:16
mithroayush3504: this is something that you'll have to look at Carlfk's and my case designs and see which ones fit better09:16
mithroayush3504: you'll probably want to do some to-scale mechanical drawings09:16
ayush3504yes. i like the daughter board idea.09:16
ayush3504can you point to those case designs?09:17
mithroayush3504: that is just one way to do the daughter boards, there are other options09:17
mithroTry looking through this -> http://www.thousandparsec.net/~irc/%23timvideos/search?q=case09:17
tpbTitle: Search IRC logs (at www.thousandparsec.net)09:17
ayush3504too many cases :P can you recall the date/month when this discussion happened09:19
ayush3504mithro: got it09:19
mithroayush3504: try maybe "box" too09:20
mithroor just search for anything CarlFK said09:21
*** slomo has joined #timvideos09:21
*** slomo_ has joined #timvideos09:24
ayush3504mithro: i just found a small discussion about using the plastic case it's shipped in. Was there something else?09:25
*** slomo has quit IRC09:27
mithroayush3504: yes I think so09:29
mithroayush3504: also check the mailing list09:29
ayush3504mithro: did you consider 3d printing the cases? it's not cheap at production level through09:32
mithroayush3504: we mostly looked at building it out of laser cut / milled parts09:32
ayush3504mithro: ok09:32
mithroayush3504: it is also about protecting the board so that inserting / removing cables doesn't destory it09:33
ayush3504mithro: yes, i do feel the need. The USB connectors on this board are quite delicate abd they are not the through-hole ones that we see in our cell-phones.09:34
mithroshenki: ping? - Do you know if it is possible to emulate a USB hub with something like the cypress controller?09:34
*** slomo_ has quit IRC09:36
mithroayush3504: ping?09:39
ayush3504mithro: yes, i'm there09:39
ayush3504mithro: i wish we could have a USB 2.0 transceiver and the rest of the USB built into FPGA. Afaik there were issues with USB 3.0 - the available pins were less and the BGA soldering was tough. Correct me if wrong.09:41
mithroayush3504: the issue with USB 3.0 is that we ran out of pins on the VHDCI connector09:41
ayush3504ayush3504: do you think it beneficial to try that on USB 2.0 instead? and maybe optimize the encoder a bit to have acceptable quality.09:43
ayush3504ah :p09:43
ayush3504mithro: ^09:43
ayush3504mithro: for the question that you left in the drawing, I would say that depends on the pitch of the pins on DB9 footprint. If it's compatible with 2.54 mm headers we don't need additional foot print for the daughter board.09:47
mithroayush3504: I don't understand your question09:47
mithroayush3504: USB 2.0 just doesn't have enough bandwidth really09:47
mithroayush3504: so there is the open question on how to do detection09:48
mithroayush3504: of the daughter boards09:48
mithroayush3504: I also think we want to look into the non rx/tx pins and better understand what is going on there09:48
ayush3504mithro: ok09:49
mithroayush3504: so - at the moment how many daughter boards are we looking at?09:50
ayush3504mithro: most of those pins are free, we can use them in any way and have jumpers to disable them if a standard serial device needs to be connected on the port. This should give you an idea - http://alex-exe.ru/images/stories/radio/microcontrollers/rs232uart/rs232uart_schem.GIF09:51
mithroayush3504: why are most of the serial pins free?09:52
ayush3504mithro: I think we can have one for RJ-45 and GPIO. I plan to look at the DMX later.09:52
*** slomo has joined #timvideos09:52
mithroayush3504: see the daughterboard table09:54
mithroayush3504: please fill out further09:54
ayush3504mithro: maybe it's because most devices don't need hardware control and the rts, cts... lines. I think it's the modems which needed these most.09:54
mithroayush3504: okay, we should justify that in your document09:55
ayush3504mithro: based on what i could find, i do not see too many daughter boards. it's just RS-485 on RJ45 and RS232 on DB9 that would do all the PTZ camera that I mentioned there. You can have DMX with XLR3/5 and GPIO with pin header other than that09:56
mithroayush3504: okay09:56
mithroayush3504: please write this down09:56
ayush3504mithro: sure09:57
ayush3504mithro: for GPIO, the 2.54mm headers look fine to me. Phoenix contacts will be expensive.10:01
mithroayush3504: how much?10:02
mithroshenki: ping?10:03
ayush3504mithro: a 10 pin set should cost $6-7 . A 2.54mm pin header is well under $110:06
mithroayush3504: okay10:06
mithroayush3504: if we are doing the daughter board approach we can go both I think?10:06
*** slomo has quit IRC10:07
ayush3504mithro: yes.10:07
ayush3504mithro: the width of these daughter boards can be much wider than the db9 foot prints. will have to consider sideways space constraints.10:09
mithroayush3504: yes, that is why the table is useful, you can figure out how wide the daughter board is by looking at the width of the connectors you are adding10:10
mithroayush3504: oh! what about infra red stuff?10:11
mithroayush3504: that could be another daughter board option?10:11
ayush3504mithro: yes10:13
ayush3504mithro: i didn't understand the input and output signal columns well10:14
mithroayush3504: if you are going to have input signals be RS232 with a RS232->RS422 converter then you are going to have input of TX/RX and output of TX-/TX+ RX-/RX+, etc10:15
ayush3504mithro: okay10:16
ayush3504mithro: and what's midi?10:16
tpbTitle: MIDI - Wikipedia, the free encyclopedia (at en.wikipedia.org)10:20
*** slomo has joined #timvideos10:20
ayush3504mithro: oh that thing. I remember seeing special connector for that on the sound cards in 90s.10:21
mithroayush3504: It is often uses for Tally lights for some reason10:22
ayush3504mithro: i think it's better for that to be a part of some audio acquisition / processing hardware.10:22
ayush3504mithro: really?10:23
ayush3504mithro: I'll look up how that works.10:23
*** sarwarc has quit IRC10:24
mithroayush3504: yes, they use like keydown to turn the light on and keyup to turn the light off10:25
*** sarwarc has joined #timvideos10:26
*** sarwarc has joined #timvideos10:26
ayush3504mithro: ah, okay10:26
ayush3504mithro: I wonder how I would test these if I make, esp the DMX and MIDI10:27
mithroayush3504: looking at USB specification, it looks like there is a way to multiplex multiple serial ports onto on set of endpoits10:32
mithrohttp://www.usb.org/developers/devclass_docs/usbcdc11.pdf <-- under the ISDN section they talk about both CAPI and "Multi-Channel communication device"10:34
mithrobut I'm unsure if that is useful for serial devices10:36
ayush3504mithro: good to know, but I doubt if I would be able to make that work on cypress10:37
mithroayush3504: why?10:39
ayush3504mithro: no specific reason as of now. I would be relying mostly on their documentation and they wouldn't have these things documented.10:40
ayush3504mithro: and maybe it gets complicated if we choose different baudrates of the ports?10:40
ayush3504mithro: will look up more on that some time later10:43
mithroayush3504: I think it's just about multiplexing the data with an ID10:43
mithroayush3504: makes life harder if the kernel driver doesn't support anything like that though10:44
ayush3504mithro: okay10:48
*** slomo has quit IRC10:48
*** Techman83-laptop has joined #timvideos10:53
mithrohttps://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/class/cdc-acm.c#n1057 looks like there isn't any type of loop10:54
tpbTitle: kernel/git/torvalds/linux.git - Linux kernel source tree (at git.kernel.org)10:54
Techman83-laptopIs there a current state of play in regards to loading the HDMI2USB firmware10:55
Techman83-laptopNo speaker tonight at PLUG, so figured I'd have a play10:55
*** slomo has joined #timvideos10:57
mithroTechman83-laptop: the libfpgalink method seems to be the most reliable10:59
Techman83-laptopmithro: cool, have we got a simple setup guide yet? I've yet to have a crack at loading firmware on it11:03
tpbTitle: libFPGALink · timvideos/HDMI2USB Wiki · GitHub (at github.com)11:04
mithroayush3504: you still here?11:04
Techman83-laptopbah, I lack the correct usb cable and the normal gear isn't here11:10
Techman83-laptopOh well, I'll build the toolset whilst I'm here before heading home11:10
mithroayush3504: poke.....11:11
mithrorohitksingh: ping? Can you reply to the Expansion board auto-detection email with your extra info?11:13
*** Techman83-laptop has quit IRC11:18
*** slomo_ has joined #timvideos11:25
*** slomo has quit IRC11:25
ayush3504mithro: back11:26
mithroayush3504: you really need to figure out the endpoint situation and if that is really a restriction on the number of serial ports11:30
mithroayush3504: can you fill in the endpoint table more fully11:30
ayush3504yes, endpoints is the restriction. The EPs I did not mentioned are free apparently.11:31
*** slomo_ has quit IRC11:31
ayush3504mithro: like the EP1 you asked11:31
mithroayush3504: I don't think endpoints are necessarily a restriction - and as I pointed out there are a bunch of other proposal which would consume endpoints11:32
ayush3504mithro: yes, looking at the doc11:33
mithroayush3504: as well, there are multiple usb-serial devices which support more then one serial port per endpoint -> see https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/serial/usb-serial.c11:34
tpbTitle: kernel/git/torvalds/linux.git - Linux kernel source tree (at git.kernel.org)11:34
ayush3504mithro: MAX_NUM_PORTS is 8 in the header file -> https://www.kernel.org/pub/linux/kernel/people/marcelo/linux-2.4/drivers/usb/serial/usb-serial.h11:40
mithroayush3504: 2.4 is *ancient*11:42
*** slomo has joined #timvideos11:45
*** slomo has joined #timvideos11:45
*** rohitksingh1 has joined #timvideos11:48
ayush3504mithro: ok, but even in ver 3.15 rc it's set to 8. -> http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/include/linux/usb/serial.h11:50
tpbTitle: Linux/include/linux/usb/serial.h (at tomoyo.sourceforge.jp)11:50
mithroayush3504: okay, so that probably the limit?11:50
mithroayush3504: I'd ask shenki a bit more about it11:50
ayush3504mithro: yes, seems so11:51
*** rohitksingh has quit IRC11:51
mithroayush3504: okay - so you might have two limiting factors then11:53
mithroayush3504: please document it11:53
ayush3504mithro: ok11:56
mithroayush3504: okay, so with your requirements document I see us needing to do the following things12:00
mithroayush3504: a) figure out the maximum number of serial ports it is possible to support12:00
mithroayush3504: b) figure out the daughter board design12:00
mithroayush3504: c) figure out if we just need RX/TX or if we really do want the remaining signals12:01
mithroayush3504: d) figure out the power supply12:01
mithroayush3504: IE you haven't mentioned the fact that RS232 needs +12V -> -12V12:02
ayush3504mithro: noted12:02
mithroayush3504: e) figure out the auto-detection methods / theory12:03
mithroayush3504: the daughter board design is going to be driven via (c), (e) and the connector size12:03
mithroayush3504: so you need to do a table which looks at the mechanical sizes of the connectors12:03
mithroayush3504: (a) is going to determined the motherboard size right?12:04
mithroayush3504: and of course (b)12:05
mithroayush3504: with the alternative way the daughter boards are designed, we possibly wouldn't need to worry about just mapping to an RS232 connector12:05
*** slomo_ has joined #timvideos12:08
*** slomo has quit IRC12:10
ayush3504mithro: you mean the daughter boards can be soldered directly or they can have a db9 connector to become an adapter as i initially thought right?12:10
mithroayush3504: I just had the idea that if the pins for interfacing between the motherboard and the daughterboard had the same layout as a female DB9 connector you could choose to use the daughterboard as stand alone, or on the motherboard?12:12
mithroayush3504: does that make sense?12:16
ayush3504mithro: sounds good. could there be a problem of RX,TX getting flipped?12:16
mithroayush3504: not sure, you'd have to make sure :P12:16
ayush3504mithro: it's hard for me to visualize, will have to doodle a bit on this12:16
mithroayush3504 / rohitksingh1: I've extended the Expansion Board spreadsheet quite a bit12:17
mithroayush3504: can I get you looking into adding the mechanical information into the spreadsheet?12:17
mithroayush3504: as you are concerned with the mechanical stuff for the serial module most12:17
mithroayush3504 / rohitksingh1: we should populate the first page further too12:20
ayush3504mithro: sure. can you tell me once what all spreadsheets to make?12:20
mithroayush3504: as many as you need to reasonably explore all the above requirements12:21
mithroayush3504: at a minimum I think you are looking at12:21
mithroayush3504: something tracking connector size + price (to inform daughter board sizing)12:21
mithroayush3504: something tracking power usage12:21
mithroayush3504: something tracking the endpoint information, as that'll inform the maximum number of serial ports12:22
mithroayush3504: some time of BOM / pricing calculation sheet to allow tracking of parts you've potentially selected and how it effects the price12:23
mithroayush3504: you should be able to tell me almost instantly how various changes in the parts change the final price outcome12:23
ayush3504mithro: sure, thanks12:24
mithroayush3504: they could just be different sheets in the same spreadsheet too btw12:24
*** slomo_ has quit IRC12:25
mithroayush3504: the whole IC selection stuff should probably be a spreadsheet where you can do things like change the line driver you are selecting12:25
mithroand change the DC<->DC converter12:26
ayush3504mithro: yes, i'll figure out a structure if it makes sense12:26
mithroayush3504: probably your daughter board stuff should be a spreadsheet too12:27
mithroayush3504: IE, it would be good to know what the prices of the various daughterboards would be12:28
mithroayush3504: so it feels pretty reasonable to me that we should be able to have a requirements document finished by this time tomorrow12:29
mithroayush3504: do you agree?12:30
ayush3504mithro: no, it'll take me one more day. I have to go somewhere today evening.12:31
mithroayush3504: okay, so we are saying COB Wednesday then? (basically when I get in on Thursday)12:32
mithroayush3504: lets try really hard to make that deadline12:33
mithroayush3504: I'll be around tomorrow night time to answer questions12:33
mithroayush3504: if you get stuck, please ask right away12:33
ayush3504mithro: sure. thanks a lot!12:34
*** slomo has joined #timvideos12:38
*** slomo has quit IRC12:43
*** rohitksingh1 is now known as rohitksingh12:49
*** slomo has joined #timvideos12:51
*** slomo has joined #timvideos12:51
rohitksinghmithro & ayush3504: Yeah! will update the spreadsheet and the will also mail about the auto-detection info12:52
mithrorohitksingh: I have a reply that I want to send out, but want you to reply first12:53
rohitksinghokay! replying it within 10 minutes now12:54
*** sarwarc has quit IRC12:59
*** sarwarc has joined #timvideos12:59
*** slomo has quit IRC13:13
*** sarwarc has quit IRC13:13
mithro~seen mparuszewski13:25
tpbmithro: mparuszewski was last seen in #timvideos 5 days, 5 hours, 29 minutes, and 32 seconds ago: <mparuszewski> Just got an email about the pull request, checked links, seems OK and merged. :D13:25
*** CarlFK has joined #timvideos13:29
*** ChanServ sets mode: +v CarlFK13:29
*** slomo has joined #timvideos13:29
*** Niharika has joined #timvideos13:37
*** slomo has quit IRC13:43
*** sarwarc has joined #timvideos13:47
*** rohitksingh has quit IRC13:49
*** rohitksingh has joined #timvideos13:55
*** slomo has joined #timvideos14:09
*** slomo has joined #timvideos14:09
*** rohitksingh has quit IRC14:19
*** slomo has quit IRC14:38
*** slomo has joined #timvideos14:39
*** mparuszewski has joined #timvideos14:49
mparuszewskimithro: Are you online? :)14:50
mparuszewskiI wanted to ask you when should I create pull request for code.timvideos.us?14:51
mithromparuszewski: shoot15:00
mithromparuszewski: regarding the planet15:00
mithromparuszewski: I think we should have a structure something like /{category/$categoryname,tag/$tagname}/{index.html, iframe.html, opml.xml, data.json}15:01
*** sarwarc has quit IRC15:01
mparuszewskimithro: hmm, for example: /gsoc-2014/news/iframe.html ?15:03
*** slomo has quit IRC15:04
mparuszewskiWe can have /gsoc-2014/{ index.html, iframe.html, etc.} and /gsoc-2014/news/ and also /news/, I can generate all available variants.15:05
mparuszewskiFirst grouped by category and then for each tag, that is possible with jekyll :)15:06
mparuszewskiI am not sure how far we can go with custom routes, I know that jekyll supports custom urls, but not every plugin will work properly if every page will have custom route. I can generate category and tag pages/feeds with structure you've written above.15:07
mithromparuszewski: no /category/gsoc-2014/iframe.html15:09
mparuszewskiOK, I understand, just like in RESTful APIs?15:10
mithromparuszewski: yeah, I guess?15:13
mparuszewskiI will try to write generator, that wouldn't be hard, I haven't seen that in Jekyll before, but it will be a challenge.15:14
*** rohitksingh has joined #timvideos15:15
mparuszewskiI think I've answered all your questions on Producteev, if you have more ideas or task feel free to add them to Producteev. I will fix now my generators to suit your needs and I will add social buttons for publishing news on FB, Twitter and G+.15:16
mithromparuszewski: social buttons?15:16
mithromparuszewski: the idea would be to publish the news without any interaction on our part15:16
mithromparuszewski: the categories / tags are just filtered versions of the primary site, right?15:17
mparuszewskiYes. :)15:18
mparuszewskiShould visitors have ability to publish news from our site to their walls?15:19
mithromparuszewski: great! that should make it easier no? And when we extend the site, each of the categories / tag subsections gets new features too?15:19
mithrohttps://github.com/LawrenceWoodman/related_posts-jekyll_plugin <-- is this useful?15:20
tpbTitle: LawrenceWoodman/related_posts-jekyll_plugin · GitHub (at github.com)15:20
mparuszewskiYes, they are configurable. Nothing is hardcoded, after the development every changes that should be made is the content, all features should work even if content has changed.15:21
mparuszewskiRelated posts are already built in jekyll, we can use it or use some of available plugins, I've seen few of them.15:22
mithromparuszewski: https://gist.github.com/enaeher/88cda643aa7e3b0ca1e5 <-- that might be useful?15:28
tpbTitle: Tiered Archives plugin (at gist.github.com)15:28
mithromparuszewski: we want to do something similar right?15:28
mparuszewskiOh thanks, yes, we want to do something similar, I will take a look on that plugin, I hope it will work with newest Jekyll, but I am sure that I will find something si,illar. :)15:30
mithrohttps://github.com/pattex/jekyll-tagging <-- we should also do this :P15:33
tpbTitle: pattex/jekyll-tagging · GitHub (at github.com)15:33
mparuszewskiYes, that's what we've been looking for! :D15:36
mithromparuszewski: this might be cool to add <-- https://gist.github.com/zachleat/579268115:36
tpbTitle: Read this in X minutes Liquid Filter Plugin (for Jekyll) (at gist.github.com)15:36
mithromparuszewski: take a look at this -> https://github.com/recurser/jekyll-plugins15:37
tpbTitle: recurser/jekyll-plugins · GitHub (at github.com)15:37
mparuszewskiYeah, simply and quite cool :)15:37
tpbTitle: recurser/jekyll-plugins · GitHub (at github.com)15:37
mithrogenerating a sitemap would probably be useful too15:39
mparuszewskiI've saved all links you've sent me. I know that octopress has also nice features, I will look at its repository. Ok, I will create new tasks on Producteev to make sure I will not forget about them.15:40
mparuszewskiThis week I would like to write new deploy script in pure Ruby without bash to make sure that page generation succeeds.15:41
mithro--> https://github.com/yihangho/emoji-for-jekyll :P15:41
tpbTitle: yihangho/emoji-for-jekyll · GitHub (at github.com)15:41
mithromaybe we want search too? -> https://github.com/PascalW/jekyll_indextank15:42
tpbTitle: pascalw/jekyll_indextank · GitHub (at github.com)15:42
mithromparuszewski: I wanted to talk to you about that15:43
mithromparuszewski: the deploy script that is15:43
mithromparuszewski: cron already has support for sending emails15:44
mithromparuszewski: (well on the output)15:44
mparuszewskiI didn't know that, I can see now, we can send email on error.15:45
mithromparuszewski: that even covers issues like the bash script fails to parse15:47
mithromparuszewski: the problem with sending the email from inside the generation script is that if the generation script fails catastrophically (IE someone commits invalid ruby) then no message will be sent15:47
mithromparuszewski: the whole "who watches the watchers" type issue :P15:48
mparuszewskiBut where should we send emails?15:48
mithromparuszewski: that is a good question :P15:48
mparuszewskiMaybe we can write on  IRC if something went wrong?15:49
mithromparuszewski: that would be good, but I don't think cron can do that....15:49
mithromparuszewski: so, often projects have a "commit message list"15:50
mithromparuszewski: We could have a list which has the github "Services / Add Email" email15:51
mithromparuszewski: the results emails could be published there too15:51
mparuszewskiOk, I will try with that.15:53
*** Niharika has left #timvideos16:37
*** sarwarc has joined #timvideos16:38
*** CarlFK has quit IRC16:45
*** CarlFK has joined #timvideos17:27
*** ChanServ sets mode: +v CarlFK17:27
*** rohitksingh has quit IRC17:50
*** sarwarc has quit IRC17:58
*** rohitksingh has joined #timvideos18:06
*** mparuszewski has quit IRC18:31
*** slomo has joined #timvideos18:49
*** slomo has joined #timvideos18:49
*** rohitksingh has quit IRC18:55
*** slomo has quit IRC19:02
*** rohitksingh has joined #timvideos19:13
*** rohitksingh1 has joined #timvideos19:39
*** rohitksingh has quit IRC19:40
*** loadstar81 has quit IRC20:33
*** skay has quit IRC20:33
*** loadstar81 has joined #timvideos20:33
*** skay has joined #timvideos20:34
*** ChanServ sets mode: +v skay20:34
*** rohitksingh1 has quit IRC23:30

Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!