[00:03] <fsphil> well that's strange .. fldigi can decode much better with 1.5 stop bits than just 1
[00:10] defy (~defy@ip-118-90-35-121.xdsl.xnet.co.nz) joined #highaltitude.
[00:10] <defy> wow, channel has really grown since I was here last
[00:11] <natrium42> :)
[00:15] <defy> did james ever launch that trans-atlantic payload?
[00:16] <natrium42> not yet
[00:17] <defy> any dates set yet?
[00:22] <natrium42> sometime in May
[00:38] <sbasuita> hmm looks like jcoxon just forgot to write the count() function ;P
[00:38] <sbasuita> anyway gnite
[00:38] sbasuita (~sbasuita@unaffiliated/drebellion) left irc: Quit: Leaving
[00:43] ProjectCirrus (~rhspm@host86-157-40-247.range86-157.btcentralplus.com) left irc:
[00:46] ms7821 (~Mark@flat.ms) left irc: Disconnected by services
[00:46] ms7821 (~Mark@flat.ms) joined #highaltitude.
[00:53] <DanielRichman> Randomskk, it works
[00:54] <Randomskk> DanielRichman: what does?
[00:54] <DanielRichman> Randomskk, output to png
[00:54] <Randomskk> :o wow
[00:54] <Randomskk> nice going
[00:54] <Randomskk> does it save it to a png in a local folder or such?
[00:59] <DanielRichman> Randomskk, yeah its very much a proof of concept
[00:59] <DanielRichman> I just pushed it to my github
[00:59] <Randomskk> nevertheless good work
[00:59] <DanielRichman> And I know it doesn't segfault on my computer using portaudio but that's about it :P
[01:00] <Randomskk> hmm I'm still suffering from the missing count() function
[01:00] <Randomskk> what've you done about that?
[01:01] <DanielRichman> commented that line out
[01:01] <DanielRichman> because it's inside an if (extracting) statement
[01:01] defy (~defy@ip-118-90-35-121.xdsl.xnet.co.nz) left irc: Read error: Connection reset by peer
[01:01] <DanielRichman> and I don't intend to turn extracting on :P
[01:01] <DanielRichman> I'll send you a pull request for the libpng hacking
[01:02] <Randomskk> no need
[01:02] <Randomskk> already merged it
[01:02] <Randomskk> pull requests are essentially just fancy messages saying "I have some new code", saying it on irc serves the same purpose
[01:02] <DanielRichman> he's too quick!
[01:02] <DanielRichman> indeed
[01:02] <Randomskk> works lovely here
[01:02] <Randomskk> gj!
[01:02] <DanielRichman> :P
[01:03] <DanielRichman> it's set to dump the png every half-rotation of the waterfall
[01:03] <DanielRichman> and it's not particularly efficient
[01:03] <DanielRichman> (copies the entire image data from one part of the memory to another then writes through libpng)
[01:03] <Randomskk> I doubt strict efficiency is the biggest concern
[01:03] <Randomskk> looks great
[01:03] Action: Randomskk has not looked at code yet
[01:04] <DanielRichman> I'll send jcoxon a pull request because I doubt he'll be back online before I go to sleep
[01:08] <Randomskk> DanielRichman: http://pastie.org/879159
[01:10] <natrium42> oh, are you adding http control to fldigi?
[01:10] <Randomskk> natrium42: hopefully yes
[01:11] <natrium42> awesome
[01:11] <Randomskk> jcoxon was of the impression that the http control would be easy but extracting the waterfall quite difficult
[01:11] <Randomskk> DanielRichman has just proved him quite wrong
[01:13] <DanielRichman> Randomskk, in terms of "efficiency" it might be better to pipe (via FIFO or otherwise) the pngs to a C program which acts as a small http server
[01:13] <DanielRichman> which just caches it in ram
[01:13] <Randomskk> possibly but I doubt it makes a significant difference
[01:13] <DanielRichman> alternatively might be easier to place the png file in /dev/shm
[01:13] <Randomskk> there's code efficiency and programmer efficiency and sometimes one is more important
[01:13] <Randomskk> hmm perhaps
[01:14] <Randomskk> but that's not cross platform
[01:14] <DanielRichman> indeed
[01:14] <Randomskk> making a small c program act as a cross platform http server is not going to be fun either
[01:14] <Randomskk> ideally this http server would be in something naturally cross platform
[01:14] <natrium42> Randomskk, do they use GTK?
[01:14] <Randomskk> e.g. python can do http and xml-rpc no problem
[01:14] <Randomskk> natrium42: no
[01:14] <Randomskk> fltk I think
[01:14] <natrium42> ah
[01:15] <Randomskk> yes, FLTK
[01:20] Enceladus (~Brian@host86-155-6-161.range86-155.btcentralplus.com) left irc: Quit: AmigaOS 4.1. Remember when computing was fun?
[01:26] <Randomskk> google app engine's data store looks really nice
[01:27] <Randomskk> much nicer than sqlalchemy imo
[01:27] <Randomskk> shame it's not available as a separate program per se really
[01:29] <DanielRichman> it has its drawbacks though
[01:29] <DanielRichman> I've been playing with it
[01:29] <Randomskk> such as? I'm still just reading through its docs
[01:29] <Randomskk> I had a play with sqlalchemy via the pylons tutorial and it just seems a bit messy
[01:29] <Randomskk> initially I didn't like GQL much but the alternative interface to it seems quite good
[01:30] <DanielRichman> I decided that I'd like free hosting from google so I wrote a quick CMS (read- store flat files in the datastore and bring them up) for my newly google-apps'd domain http://www.danielrichman.co.uk/
[01:30] <Randomskk> I am having a similar conflict, in that I already have a good web server but google app engine has other advantages and so does heroku for ruby
[01:30] <Randomskk> atm I really can't decide which I prefer, ruby or python
[01:30] <Randomskk> need more experience with python web apps and ruby in general first I think
[01:31] <Randomskk> activerecord combined with ruby's syntactical sugar is delicious, but I'm not a massive fan of model properties being defined in migrations
[01:31] <Randomskk> otoh django's ORM wasn't actually that bad thinking about it, but I generally didn't love django (despite its excellent docs) - not sure about now though
[01:32] <DanielRichman> ahhh, Models
[01:32] <Randomskk> I like django's templating well enough but rail's activetemplate is also good
[01:33] <Randomskk> rails in general would be excellent but has a few sticking points like the whole migrations being the only record of model's properties
[01:33] <DanielRichman> django templates are very nice. I was using the GAE models and sometimes the features that make you think "yeah, that'll be nice" come back to bite you
[01:33] <Randomskk> any examples?
[01:33] <DanielRichman> eg. I setup required=True, validator=function on several properties of my models, and then wanted the simple code of
[01:33] <DanielRichman> 1) get Object from datastore if exists or create it
[01:33] <DanielRichman> 2) get values from self.request.get() and assign them to object
[01:34] <DanielRichman> 3) then validate object, if OK put() it; if not then re-display the edit page with an error message
[01:34] <DanielRichman> (but keep previous values; ie. don't discard any work even if invalid)
[01:34] <Randomskk> fairly standard pattern
[01:34] <DanielRichman> but the Models just adamantly refused to have any invalid values assigned to them
[01:34] <Randomskk> they do
[01:35] <Randomskk> hmm, I did read about that
[01:35] <Randomskk> validation occurs on value assignment etc
[01:35] <DanielRichman> I was probably Doing It Wrong (tm) but if I wanted to work around I'd have to build a dict separate from the Object in order to pass those values back to the template
[01:35] <Randomskk> I imagine there's nevertheless a good way to get that pattern
[01:35] <Randomskk> eh
[01:35] <Randomskk> what you should be able to do is some kind of Form class
[01:35] <DanielRichman> yeah
[01:35] <Randomskk> would be the typical way to handle it
[01:35] <Randomskk> then the Form class renders itself into pretty html
[01:35] <DanielRichman> I have heard about them but didn't look into it
[01:36] <Randomskk> also you can set errors against fields on the Form class and they show up in the right place
[01:36] <Randomskk> and it would pass you out a dict of the data needed ready to be set into the model
[01:36] <Randomskk> but also maintain it internally
[01:36] <Randomskk> http://docs.djangoproject.com/en/1.1/topics/forms/
[01:37] <Randomskk> or in this case rather
[01:37] <Randomskk> http://docs.djangoproject.com/en/1.1/topics/forms/modelforms
[01:37] <DanielRichman> it seems that GAE does provide forms
[01:37] <Randomskk> is more relevant
[01:37] <Randomskk> modelforms are what you want here
[01:37] <Randomskk> modelforms are super useful
[01:38] <DanielRichman> Also (Was again probably doing it wrong) but the Datastore run_in_transaction refused to run queries (?)
[01:38] <Randomskk> what were you trying to transaction?
[01:38] <DanielRichman> ie you can use it to atomically .get(), increase ++ a property, and .put()
[01:38] <DanielRichman> but I wanted to create or update
[01:38] <Randomskk> yea
[01:38] <DanielRichman> GAE provides a function to do that but it seems to want a key_name
[01:38] <Randomskk> there are caveats like a transaction with multiple Models has to be told in advance that those models will be transactioned together
[01:39] <DanielRichman> rather than provide uniqueness based on one of my string properties
[01:39] <Randomskk> hard to say if there's a way around that or not, would have to look into it
[01:39] <DanielRichman> yep
[01:39] <Randomskk> generally I understand that google's transaction support is quite good though
[01:39] <Randomskk> what form support do google provide?
[01:39] <DanielRichman> I think it's django standard
[01:39] <Randomskk> they just provide the django forms?
[01:40] <fsphil> night guys
[01:40] <Randomskk> that'd be nice but dunno if we'd get modelforms then
[01:40] fsphil (~phil@2001:470:1f09:483:21f:c6ff:fe44:b25b) left irc: Quit: Ka-Boom!!!!
[01:40] <Randomskk> since that uses django's orm
[01:40] <DanielRichman> daniel@desktop:appengine-sdk$ tree
[01:40] <DanielRichman> ...
[01:40] <DanielRichman> ./lib/django/django/forms
[01:40] <Randomskk> curious
[01:40] <DanielRichman> ./lib/django/django/newforms
[01:40] <Randomskk> newforms are what you generally want
[01:40] <Randomskk> curious
[01:40] <DanielRichman> and also
[01:40] <DanielRichman> ./google/appengine/ext/db/djangoforms.py
[01:40] <Randomskk> exciting
[01:40] <Randomskk> that could be good
[01:40] <Randomskk> tempting to make an app in pure django and then later worry about delpoying it to GAE
[01:40] <Randomskk> rather than writing for GAE off the bat
[01:41] <Randomskk> dunno how google data store would integrate with django's ORM though
[01:41] <Randomskk> "The Django data modeling interface is not compatible with the App Engine datastore. You can use the App Engine data modeling libraries in your Django applications."
[01:41] <DanielRichman> I think there is a project out there that tries to patch it in, but you definitely want to double check that
[01:41] <DanielRichman> http://code.google.com/appengine/articles/djangoforms.html
[01:42] <Randomskk> sadly that buggers cross-host compatibility
[01:42] <Randomskk> can't write GAE code and host it myself
[01:42] <Randomskk> can't write django ORM code and host it on google
[01:42] <DanielRichman> You could pimp ./dev_appserver.py :P
[01:42] <Randomskk> hmm
[01:42] <Randomskk> well I mean, it's WSGI-compliant
[01:43] <Randomskk> would work with any python webserver that does WSGI
[01:43] <Randomskk> using their SDK datastore
[01:43] <Randomskk> but it's not going to scale at all
[01:43] <Randomskk> there might be a good way to host your own version though, interesting thought
[01:43] <DanielRichman> well the dev_appserver is, a dev server, and is immensely inefficient.
[01:44] <DanielRichman> it would need some serious work
[01:44] <Randomskk> yea, but all you'd need is a db implementation
[01:44] <Randomskk> you can use a decent python server
[01:44] <DanielRichman> true
[01:44] <Randomskk> http://code.google.com/appengine/articles/pure_django.html hmm
[01:45] <Randomskk> apparently that helper program will map google db related things onto an SQL database
[01:45] <Randomskk> "GQL queries are analyzed and translated into Django field lookups. The returned QuerySets work almost exactly the same as result sets in App Engine."
[01:45] <Randomskk> clever
[01:46] <DanielRichman> nice
[01:46] <Randomskk> http://code.google.com/appengine/articles/app-engine-patch.html too, goes the other way
[01:46] <DanielRichman> Oh and beware the pre-django 1.0
[01:46] <Randomskk> "app-engine-patch is a project that aims for providing all the functionality of Django while working around the limitations imposed by the missing Django ORM support. "
[01:48] <Randomskk> includes having things like authenticating non-google users
[01:48] <Randomskk> the easy tie-in to many google services is quite tempting too
[01:48] <Randomskk> like mail
[01:48] <Randomskk> on the other hand, heroku remains a more awesome app platform in my opinion - deply by git pushing!
[01:48] <Randomskk> deploy* even
[01:49] <DanielRichman> Randomskk, remember that GAE runs on django 0.98
[01:49] <Randomskk> it can run on 1.0 and 1.1 too though
[01:49] <Randomskk> you just make a call to the api to tell it to use the new version before importing django
[01:50] <DanielRichman> oh?
[01:50] <Randomskk> http://code.google.com/appengine/docs/python/tools/libraries.html
[01:50] <Randomskk> from google.appengine.dist import use_library
[01:50] <Randomskk> use_library('django', '1.1')
[01:51] <DanielRichman> But does that then integrate with the google.appengine.api.webapp shortcuts?
[01:52] <Randomskk> apparently
[01:53] <DanielRichman> also on GAE you get free https (only on .appspot.com though)
[01:53] <Randomskk> eh, I have free https anyway though
[01:53] <Randomskk> not a massive concern
[01:53] <DanielRichman> with a full/proper certificate?
[01:54] <Randomskk> yea
[01:54] <Randomskk> https://randomskk.net/
[01:55] <Randomskk> sweet, adamgreig.com is top of google for "adam greig" again
[01:55] <DanielRichman> and that means you must have a dedicated server?
[01:55] <Randomskk> well I have a VPS with two IPs
[01:55] <Randomskk> I did have a dedicated server for like a year
[01:55] <Randomskk> but the cost wasn't really worth it compared to this vps which is actually more powerful
[01:55] <Randomskk> (slicehost.com)
[01:55] <DanielRichman> mmm it's the dedicated IP that does the trick :P
[01:55] <Randomskk> also means I get to play at being a sysadmin which is loads of fun
[01:56] <Randomskk> atm my server isnt' configured to host ruby or python though
[01:56] <Randomskk> hence some interest in GAE and heroku
[01:56] <Randomskk> heroku is just too cool and another reason I want to play with ruby more
[01:56] <Randomskk> similar reasoning with github led to me liking git, though also it's clearly awesome
[01:56] <DanielRichman> heh. I've been allowed to have a server box on 24/7 on our (fairly poor) broadband; have been hosting up a few mini projects
[01:56] <Randomskk> yea, I started off like that :P
[01:56] <DanielRichman> but am finding fewer and fewer uses for it
[01:56] <Randomskk> my isp gave us a static IP
[01:56] <DanielRichman> Yeah, I used to have one then the ISP got bought out
[01:57] <Randomskk> but then I moved to a dedi, and then to this vps
[01:57] <DanielRichman> dyndns, gogogo
[01:57] <Randomskk> https://randomskk.net/ranger/uptime/
[01:57] <Randomskk> mmm 158 days
[01:57] <DanielRichman> no kernel updates?
[01:57] <Randomskk> not in the last 158 days, no
[01:57] <DanielRichman> :P
[01:57] <Randomskk> and those other reboots were a notified-a-week-in-advance deal
[01:57] <Randomskk> slicehost have generally been awesome
[01:58] <Randomskk> would probably be worth learning to host ruby and python projects anyway but atm my web server is apache due to multiuser support
[01:58] <DanielRichman> US based datacenter?
[01:58] <Randomskk> yea
[01:58] <Randomskk> latency is fine though
[01:58] <Randomskk> heck it's running irssi and I'm using that over ssh as my main irc client
[01:58] <Randomskk> latency when typing would be killer
[01:58] <Randomskk> but it's fine
[01:58] <DanielRichman> wow, nice
[01:58] <DanielRichman> Once tried hosting python on apache; was fairly painless
[01:59] <DanielRichman> though felt a bit hacky
[01:59] <Randomskk> doing it with mod_python is painless but hacky and not very efficient
[01:59] <Randomskk> I'd like to use either nginx which is what all the cool kids use, lighty which is what some older cool kids use or cherokee which I like the look of but not many cool kids use
[01:59] <Randomskk> but none are as good as apache at securely hosting lots of different users who I don't trust
[02:00] <Randomskk> I want to move them all to something else and then run my own http for my websites but eh
[02:00] <Randomskk> some time
[02:00] <Randomskk> ideally I want them all to bugger off >.>
[02:00] <DanielRichman> ssb got a shellbox account a while back, which was lighttpd based, and it didn't take long for me to point out the PHP-running-unpatched-as-UID lighttpd
[02:00] <DanielRichman> so he told the admin and took the credit :P
[02:01] <DanielRichman> (they still haven't fixed it)
[02:01] <Randomskk> nice nice
[02:01] <Randomskk> typical
[02:01] <Randomskk> my apache installation is somewhat more secure
[02:01] <Randomskk> i.e. there is no shell access and also php runs as that website's own user with permission only to access its own files
[02:01] <Randomskk> also apache gets chrooted inside the web hosting directory so can't actually touch the system
[02:01] <DanielRichman> I think that counts as secure
[02:01] <DanielRichman> Slicehost is a Rackspace company: Strange, the rackspace dedicated box I have access to has fairly bad latency over ssh
[02:02] <Randomskk> rackspace have many datacentres
[02:02] <Randomskk> variety of other security things in place too
[02:02] <Randomskk> oh and then I run ftp
[02:02] <Randomskk> fucking ftp
[02:02] <Randomskk> fucking people can't use sftp
[02:02] <DanielRichman> oh man, ftp
[02:02] <DanielRichman> that's nasty
[02:02] <Randomskk> yea really
[02:02] <Randomskk> "but I can't update my website without ftp!!!"
[02:02] <Randomskk> "well maybe you shouldn't have a website" etc
[02:02] <Randomskk> only their accounts are allowed to log in with ftp anyway though
[02:03] <Randomskk> they are the only ones getting fucked when they do so from someone's open wifi
[02:03] <Randomskk> and all sorts of mechanisms in place to lock that down if too many bad attempts etc
[02:03] <DanielRichman> what do you do for backups? or do slicehost take care of it
[02:04] <Randomskk> daily cronjab does some rsync and cp magic to create hardlinked backups for the last week in a neat form
[02:04] <Randomskk> inside each person's directory there's a /backups
[02:04] <Randomskk> that has a folder for each day of the last week
[02:04] <Randomskk> each folder is their space as it was then
[02:04] <Randomskk> but it doesn't use any more disk space unless they change the files currently
[02:04] <Randomskk> e.g. hardlinks
[02:05] <Randomskk> so when there's been no change in the last week it uses like a few kilobytes more space
[02:05] <DanielRichman> nice, nice
[02:05] <Randomskk> and then monthly it's all compressed, encrypted and uplaoded to amazon s3
[02:05] <DanielRichman> but of course. I've not got that much important stuff so it's on an external hdd, but I was looking at s3 for a while
[02:05] <Randomskk> who charge me approx $0.20 a month to upload and store all that data
[02:05] <Randomskk> which I find a pretty good deal
[02:06] <DanielRichman> is the hardlinking a rsync option or is there a large amount of "magic" involved?
[02:06] <DanielRichman> ah wait, of course
[02:06] <Randomskk> a combination of rsync and cp magic
[02:06] <DanielRichman> Does it work if the backup target is on a different volume/hdd?
[02:06] <Randomskk> it's not that clever
[02:06] <Randomskk> don't think so.
[02:07] <Randomskk> oh, the whole system also dumps the sql database at appropriate moments to back that up too
[02:07] <DanielRichman> hmm, ok. I've wanted to have a program that will mirror / to /mnt/backup/dd-mm-yy but hardlink identical files
[02:07] <DanielRichman> rdiff-backup sortof solves the problem but it's quite slow
[02:07] <DanielRichman> and feels fragile
[02:08] <Randomskk> http://pastie.org/private/ntwkg8rlecpwq2aeztpjcq is my magic
[02:09] <Randomskk> that cp -al bit is important
[02:09] <Randomskk> looks like actually it does a full rsync backup and then cp to make the older-than-one-day backups take no space
[02:10] <Randomskk> at any rate I currently have two simple web apps I want to make, both of which are quite similar
[02:10] <Randomskk> so my plan is to do one in ruby and one in python and thus get a better idea of both
[02:10] <DanielRichman> ahh, that looks like it would work across devices
[02:10] <Randomskk> frameworks and hosts are still undecided
[02:10] <Randomskk> probably rails for ruby though sinatra is nice
[02:11] <Randomskk> but django? webapp? pylons? for python, no idea
[02:11] <Randomskk> might try both self hosting and hosting on heroku/gae too
[02:14] <DanielRichman> here's how I do my backups
[02:14] <DanielRichman> http://pastie.org/private/be6g2bu4aiugrr9mzfdgq
[02:14] <DanielRichman> but I think I'll replace the rdiff-backup section with something less ugly
[02:15] <Randomskk> have you looked into duplicity?
[02:15] <Randomskk> http://www.nongnu.org/duplicity/
[02:16] <DanielRichman> is it just rsync, tar, gpg?
[02:16] <DanielRichman> I think ssb mentioned it a while back
[02:16] <Randomskk> kinda but I think it's a bit clever and a bit easy to use
[02:17] natrium42 (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Ping timeout: 240 seconds
[02:17] <DanielRichman> hmm, ok
[02:17] <Randomskk> I can't remember why I didn't like django now
[02:17] <Randomskk> tempted to go with that for this app but meh gae
[02:17] <Randomskk> webapp would probably be powerful enough for what I want anyway, too
[02:19] natrium42 (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[02:26] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Quit: The cake is a lie !
[02:28] <Randomskk> right sleep, seeya all
[02:42] DanielRichman (~DanielRic@unaffiliated/danielrichman) left irc: Remote host closed the connection
[02:48] <juxta> anybody still awake? :)_
[03:08] natrium_ (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[03:35] natrium_ (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Quit: My other car is a cdr.
[03:36] natrium_ (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[05:52] natrium42 (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: Connection reset by peer
[06:04] juxta_ (~blah@ joined #highaltitude.
[06:18] MoALTz (~heh@ left irc: Quit: Leaving
[06:22] darknesslord_ (~darknessl@ joined #highaltitude.
[06:31] MoALTz (~heh@ joined #highaltitude.
[06:46] juxta_ (~blah@ left irc: Ping timeout: 276 seconds
[07:04] _sh3 (debian-tor@gateway/tor-sasl/sh3/x-62271040) left irc: Remote host closed the connection
[07:06] _sh3 (debian-tor@gateway/tor-sasl/sh3/x-62271040) joined #highaltitude.
[07:11] darknesslord_ (~darknessl@ left irc: Quit: darknesslord_
[07:28] Simon-MPFH (~simon@phantom.mpfh.co.uk) joined #highaltitude.
[07:41] wasutton3-1 (~will@ left irc: Ping timeout: 268 seconds
[07:46] natrium_ (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Quit: My other car is a cdr.
[07:46] wasutton3-1 (~will@ joined #highaltitude.
[08:01] jonsowman (~jonsowman@93-97-184-163.zone5.bethere.co.uk) left irc: Ping timeout: 246 seconds
[08:02] jonsowman (~jonsowman@93-97-184-163.zone5.bethere.co.uk) joined #highaltitude.
[08:07] edmoore (~ed@pluto.trinhall.cam.ac.uk) joined #highaltitude.
[08:47] edmoore (~ed@pluto.trinhall.cam.ac.uk) left irc: Ping timeout: 264 seconds
[09:14] edmoore (~ed@chu-gw-a.churchillcambridge.co.uk) joined #highaltitude.
[09:21] edmoore (~ed@chu-gw-a.churchillcambridge.co.uk) left irc: Quit: edmoore
[09:25] jcoxon (~jcoxon@ joined #highaltitude.
[09:25] <juxta> hi all
[09:25] <jcoxon> morning juxta
[09:25] <juxta> hey jcoxon
[09:25] <juxta> how's things?
[09:27] <jcoxon> good thanks, in the library
[09:27] <jcoxon> you?
[09:28] <juxta> not too bad, sunday night here
[09:28] <juxta> trying to work out why the UKHAS list keeps unsubscribing me :(
[09:29] <jcoxon> thats odd
[09:30] <juxta> jcoxon: any idea if RTTY in the new version of fldigi is much of an improvment?
[09:31] <jcoxon> apparently
[09:31] <jcoxon> decodes the wav files i have just fine :-p
[09:31] <juxta> ones that the old version wont?
[09:32] <jcoxon> it decodes them just fine as well
[09:32] <jcoxon> hehe
[09:32] <juxta> oh, haha
[09:32] <jcoxon> the good thing is the new fldigi has custom shift
[09:32] <juxta> yeah
[09:32] <juxta> that's nice
[09:32] <jcoxon> the rtty decoder won't be worse
[09:32] <juxta> would be good if it did autp-bauding too
[09:33] <jcoxon> so it'll be worth shifting to the new dl-fldigi when i get it actually working
[09:34] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Quit: Leaving.
[09:34] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[09:35] LazyLeopard (~irc-clien@chocky.demon.co.uk) joined #highaltitude.
[09:36] <jcoxon> much sunnier today :-)
[09:38] <LazyLeopard> --> Bluebell Railway, I think... ;)
[09:38] LazyLeopard (~irc-clien@chocky.demon.co.uk) left irc: Client Quit
[09:39] <jcoxon> i went there as a child :-)
[09:42] <jcoxon> wow danielrichmans png hack is impressively simple and works really quite well
[09:42] <juxta> for grabbing the waterfall?
[09:43] <jcoxon> yeah
[09:52] <Randomskk> jcoxon: looks like the web control will be the harder part after all :P
[09:52] <Randomskk> should be fairly straightforward to detect click location on the waterfall and snap the decoding bars to there
[09:52] <Randomskk> wondering if it'd be easier to have the web server integrated into dl-fldigi or not
[09:54] <Randomskk> though I'd be a bit concerned writing a web server in c++ for something like this, way too much potential to make a small mistake that leads to an exploitable buffer overflow or something
[09:54] <jcoxon> Randomskk, oh xml-rpc can do all that
[09:54] <Randomskk> would be the easiest to run though
[09:55] <Randomskk> jcoxon: do all which?
[09:55] <jcoxon> it can let you set decoder lines
[09:55] <Randomskk> excellent
[09:55] <jcoxon> will tell you signal strength
[09:55] <Randomskk> even so an integrated web server would probably be easiest ro use
[09:55] <Randomskk> to use*
[09:55] <Randomskk> I guess a python app will be my first choice then
[09:55] <jcoxon> yeah
[09:55] <Randomskk> xml-rpc plus some kind of simple http server
[09:55] <jcoxon> or php i guess
[09:56] <Randomskk> php requires something else like apache or lighty
[09:56] <jcoxon> well this option isn't for the faint hearted
[09:56] <Randomskk> which then requires installation and a suitable php decoder to be installed too
[09:56] <Randomskk> a python app can be bundled into a single exe
[09:56] <Randomskk> with all required libs and the python interpreter included
[09:56] <jcoxon> Randomskk, so python + sockets to connect?
[09:56] <Randomskk> yea
[09:57] <jcoxon> i'm more in favour of the browser based approach
[09:57] <Randomskk> it'd still be browser based
[09:57] <Randomskk> http server
[09:58] <Randomskk> http://docs.python.org/library/basehttpserver.html
[09:58] <jcoxon> oh right, so the remote station has dl-fldigi + python app running
[09:58] <Randomskk> yea
[09:58] <jcoxon> and then you connected over the net
[09:58] <jcoxon> fair enough
[09:58] <Randomskk> python can talk to xml-rpc easily
[09:58] <jcoxon> howabout for dl-fldigi we have a tab on config for "Remote"
[09:58] <Randomskk> and also act as a simple web server easily
[09:58] <Randomskk> sounds good
[09:58] <jcoxon> and in there you can turn waterfall png on and off
[09:59] <Randomskk> yup
[09:59] <Randomskk> well
[09:59] <Randomskk> why not make it an xml-rpc thing
[09:59] <Randomskk> enable/disable waterfall production and returns the URI of the generated image
[09:59] <jcoxon> thats true
[09:59] <Randomskk> dunno how easy it'd be to add an xml-rpc capability but presumably not too hard
[10:00] <jcoxon> it should be pretty simple
[10:00] <jcoxon> fldigi has a global config system
[10:00] <Randomskk> what kinda update rate should we aim for? currently it's every half cycle, and
[10:00] <jcoxon> so we can add pngwaterfall to that
[10:00] <jcoxon> its a bit fast really
[10:00] <Randomskk> http://pastie.org/879159
[10:00] <jcoxon> you'll be relying on the afc to keep track locally
[10:01] <Randomskk> shove that in a web browser and it'l update it as new images come in
[10:01] <Randomskk> the waterfall image is 13K though
[10:01] <Randomskk> it's not really very big
[10:01] <jcoxon> its something we can play with at a later date
[10:02] <Randomskk> yea
[10:02] <jcoxon> even change the refresh rate
[10:02] <Randomskk> oh true
[10:02] <Randomskk> as far as the interface goes (to start with at least) we just want what,
[10:02] <Randomskk> decoded text
[10:02] <Randomskk> mode select
[10:02] <Randomskk> frequency display/select (for hamlib)
[10:02] <Randomskk> waterfall that's clickable
[10:02] <Randomskk> and mode config?
[10:02] <Randomskk> plus I guess any HAB-info pane we add to main
[10:03] <Randomskk> by the way, dl-fldigi didn't compile as of your latest commit last night
[10:03] <Randomskk> error in fldigi-rx_extract, seems you forgot to make the count() function
[10:03] <jcoxon> really?
[10:03] <jcoxon> compiles here
[10:03] <Randomskk> is count() in the commit?
[10:04] <Randomskk> also a few warnings you could probably avoid wrt comparing signed and unsigned ints
[10:04] <Randomskk> this did give us the chance to play with another excellent git feature, bisect
[10:05] <Randomskk> `git bisect start; git bisect bad; git bisect good dab90537e; git bisect run make`
[10:05] <Randomskk> tell git to start doing a binary search for the commit that causes the error, tell it the current commit is bad and that dab90537e worked fine, then tell it to start searching, using 'make' as the command to test
[10:06] <Randomskk> git makes the inbetween repo, then if it worked the later inbetween, or if it was bad the earlier inbetween, etc
[10:06] <Randomskk> and a few seconds later tells you which commit introduced the bug
[10:06] <jcoxon> Randomskk, whats odd is that its identical to the old setup
[10:06] <Randomskk> jcoxon: it's just that line about count()
[10:07] <Randomskk> logger/rx_extract.cxx +173
[10:07] <jcoxon> yeah i'm looking at it right now
[10:08] <jcoxon> Randomskk, whats odd is that rx_extract worked for me just now when i tested teh waterfall feature
[10:08] <jcoxon> i'll make clean and make again
[10:09] <Randomskk> I don't see any definition of count()
[10:09] <Randomskk> `git status` to check there aren't uncommited files that might contain it
[10:09] <Randomskk> there's an unsigned count() in spot/pskrep and another in Fl_Native_File_Chooser_FLTK but neither seems relevant
[10:11] <Randomskk> another fun thing: adding a .gitignore file and putting "*.o" in it to make git stop saying about all the object files that aren't being tracked
[10:11] <Randomskk> that's my commit f254c7c94
[10:11] <jcoxon> compiles clean
[10:11] <Randomskk> platform differences?
[10:11] <Randomskk> where's it getting count from?
[10:12] <Randomskk> (& did you check it's not defined in some file not being included in git?)
[10:12] <jcoxon> count is a c++ function
[10:12] <Randomskk> from which library?
[10:12] <jcoxon> algorithm
[10:13] <Randomskk> http://www.cplusplus.com/reference/algorithm/count/ that one?
[10:13] <jcoxon> but it must be from somewhere else on osx i guess
[10:13] <Randomskk> well you aren't including <algorithm>
[10:13] <Randomskk> if I include algorithm it compiles fine
[10:14] <jcoxon> yeah i see - algorithm was in the original code
[10:14] <jcoxon> but i missed that bit
[10:14] <jcoxon> obviously on osx count() gets defined somewhere else as well
[10:14] <Randomskk> weird
[10:14] <Randomskk> I wonder if it's the same count()
[10:14] <Randomskk> anyway easy fix
[10:15] <Randomskk> on line 44 if you change 'int i' to 'unsigned int i' you get rid of a compiler warning
[10:15] <jcoxon> Randomskk, fix it yourself!
[10:15] <jcoxon> :-p
[10:15] <Randomskk> have done
[10:15] <Randomskk> along with the algorithm thing
[10:15] <jcoxon> oh right
[10:15] icez (~icez@unaffiliated/icez) left irc: Remote host closed the connection
[10:15] <Randomskk> but otoh it's a single word fix :P
[10:15] <jcoxon> okay
[10:15] <Randomskk> less typing to add that than merge
[10:15] <jcoxon> fair enough
[10:16] <jcoxon> i've done it on my local copy
[10:17] <jcoxon> i feel sorry for rjh when he has to fight to get this to compile on windows
[10:17] <Randomskk> ln92 could do with the same thing
[10:17] <Randomskk> int to uinsigned int
[10:17] <jcoxon> total_string_length ?
[10:17] <Randomskk> yea
[10:17] Laurenceb (~laurence@host81-154-84-43.range81-154.btcentralplus.com) joined #highaltitude.
[10:17] <Randomskk> string.length() will never be negative so returns unsigned
[10:21] <jcoxon> will get on and do the gui bits later today
[10:22] G8DSU (~chatzilla@cpc3-mort4-0-0-cust192.croy.cable.virginmedia.com) left irc: Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100202165920]
[10:26] <Randomskk> jcoxon: a little bit more code stops the warning about line 192, I can commit it if you want
[10:26] <Randomskk> have you commited those other three small changes?
[10:27] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 264 seconds
[10:28] <jcoxon> i'll do that now
[10:28] <jcoxon> done
[10:36] <Randomskk> my 1f6b3db73b fixes one more warning
[10:36] <Randomskk> bit more code
[10:37] <Randomskk> should ff to your head
[10:37] <jcoxon> so once i pull it should i then push it onto github to keep that up to date?
[10:38] <Randomskk> yup
[10:39] <jcoxon> i like dl-fldigi when it works, quite proud of it
[10:40] <Randomskk> yea
[10:40] <Randomskk> how many flights have we tracked now?
[10:40] <jcoxon> i think there are 20 flight logs on the server
[10:40] <jcoxon> rjh keeps them all stored away
[10:41] MoALTz_ (~heh@ joined #highaltitude.
[10:41] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[10:44] MoALTz (~heh@ left irc: Ping timeout: 248 seconds
[11:00] MoALTz_ (~heh@ left irc: Quit: Leaving
[11:01] MoALTz (~heh@ joined #highaltitude.
[11:04] Nick change: \mct -> mct
[11:07] sbasuita (~sbasuita@unaffiliated/drebellion) joined #highaltitude.
[11:11] <rjharrison> Hey jcoxon
[11:11] <rjharrison> I got the git working
[11:12] <rjharrison> but when i autoconf it dies with loads of errors
[11:12] <Randomskk> rjharrison: you need to `autoreconf -vfi`
[11:12] <Randomskk> note the re
[11:12] <rjharrison> Randomskk :)
[11:12] <rjharrison> Cheers
[11:12] <Randomskk> np
[11:12] <Randomskk> that is buried in docs on their website somewhere
[11:12] <rjharrison> cool
[11:12] <rjharrison> RTFM and all that
[11:12] <Randomskk> took us ages to find :P
[11:12] <Randomskk> yea quite
[11:13] <Randomskk> never seen autoreconf before but hey, it works
[11:13] <Randomskk> stupid autotools
[11:13] <rjharrison> glad i asked
[11:13] <rjharrison> I have to say this all makes dev far more awkward
[11:13] <Randomskk> rjharrison: if you go on http://github.com/jamescoxon/dl-fldigi you can click 'fork' on the top right to have github make you your own version
[11:13] <rjharrison> I understand for 50 devs
[11:13] <Randomskk> rjharrison: git or autoreconf?
[11:13] <rjharrison> but for 1 / 2
[11:13] <rjharrison> git really though I'm getting there
[11:14] <Randomskk> try it out for a bit and with some common use cases first I guess
[11:14] <Randomskk> did you see the cool thing we were able to do with it last night?
[11:14] <rjharrison> autoconf I guess makes the configure file automatically based on the source
[11:14] Jasperw (~jasperw@client-86-25-212-88.cht-bng-015.adsl.virginmedia.net) joined #highaltitude.
[11:14] <Randomskk> some commit a while ago had introduced a bug that stopped it compiling on linux despite working on mac
[11:14] <Randomskk> so jcoxon had made more commits since
[11:14] <Randomskk> git can do a binary search on the commit tree, automatically running make on each different commit, to find the commit that causes the failure
[11:15] <rjharrison> cool
[11:15] <Randomskk> you just type in the command, hit enter, git tells you which commit contains the bug
[11:15] <rjharrison> make for linux and mac environs?
[11:15] <rjharrison> or was it a gcc bug
[11:15] <rjharrison> ie not a library problem
[11:15] <Randomskk> well the bug was that he hadn't included <algorithm.
[11:15] <Randomskk> for some reason mac has count() defined always
[11:15] <rjharrison> cool
[11:15] <Randomskk> while on linux you need to include <algorithm> to get that
[11:16] <jcoxon> rjharrison, git is better then svn
[11:16] <Randomskk> equally DanielRichman was able to make a commit to his repository that has dl-fldigi output a PNG of the waterfall live, so that another app can display it
[11:16] <Randomskk> and then jcoxon is able to just pull his commit and it merges into jcoxon's repository
[11:17] <Randomskk> without having to actually set anything up wrt giving people access to svn etc
[11:17] <rjharrison> cool
[11:17] <rjharrison> So it's good for lots of devs
[11:17] <rjharrison> I'll play somemore
[11:17] <Randomskk> it's excellent for distributed projects and for open source projects
[11:18] <Randomskk> but those same properties make it very good generally - certainly it can do everything svn can do
[11:18] <Randomskk> if you wanted, you could use it in the exact same way you use svn
[11:18] <Randomskk> it just has a lot of niceties on top
[11:18] <Randomskk> rjharrison: might be worth reading http://www.eecs.harvard.edu/~cduan/technical/git/, five shortish pages that explain succintly how git works and how to use it
[11:18] <rjharrison> Do we need to be care full if we have lots of people doing stuff in lots of places on fldigi then we are going to have bigger update issues as fldigi moves on
[11:19] <rjharrison> Randomskk will do thanks for the pointer
[11:19] <rjharrison> I need it atm
[11:19] <Randomskk> rjharrison: using git makes it very easy to keep track of that
[11:19] <Randomskk> it should be simple to keep up to date with mainline fldigi
[11:19] <rjharrison> Cool thats what we want
[11:19] <jcoxon> we've already once updated to the latest upstream
[11:19] <Randomskk> for example we've already integrated three commits from mainline since starting this
[11:19] <rjharrison> So I should take out a local copy of jcoxons version
[11:20] <Randomskk> rjharrison: what you should do ideally is go to his github page and click 'fork'
[11:20] <Randomskk> it'l create a github fork of his repository on your account
[11:20] <rjharrison> jcoxon and you get to see where chnges effect our code?
[11:20] <rjharrison> ok i'll do that now
[11:20] <Randomskk> then you can "git clone git@github.com/rharrison/dl-fldigi" to get it onto your computer
[11:20] <rjharrison> jcoxon have you got the pipes working?
[11:20] <Randomskk> uh, make that
[11:21] <Randomskk> git clone git@github.com:rharrison/dl-fldigi.git
[11:21] <rjharrison> :)
[11:21] <rjharrison> cool
[11:21] <Randomskk> once you setup the fork it's worth then adding jcoxon's repository as a 'remote' so you can pull from it easily
[11:21] <rjharrison> is there a command to run to get latest from james and my git
[11:22] <Randomskk> git remote add jamescoxon git://github.com/jamescoxon/dl-fldigi.git
[11:22] <Randomskk> you do that once to add 'jamescoxon' as a remote repository
[11:22] <Randomskk> and then 'git pull jamescoxon' will get the latest from him and merge it
[11:22] <rjharrison> awsume
[11:22] <rjharrison> Ok so the git is coool
[11:22] <rjharrison> Ok so the git is cool
[11:22] Hiena (~Hiena@ joined #highaltitude.
[11:22] <rjharrison> what a git :)#
[11:23] <jcoxon> rjharrison, haven't got pipes working
[11:23] <Randomskk> probably need to do a 'git pull jamescoxon master' actually, though you can tell it to use master by default
[11:23] <jcoxon> was getting rx_extract intergrated first so that it actually functioned
[11:23] <Randomskk> rjharrison: but at any rate that website I linked you do explains how it works better than I can
[11:23] <jcoxon> was going to focus on getting the gui up to scratch a bit as its pretty unusable right now
[11:25] <rjharrison> Randomskk appreciate the 5 min spin on the git
[11:25] <rjharrison> jcoxon right I'll get the pipes pipling
[11:25] <rjharrison> piping (sp?)
[11:25] <Randomskk> give me a shout if you have any questions or don't get something and I'll see if I can explain
[11:25] <jcoxon> i've put the pipe code in a header file which is a bad idea
[11:25] <jcoxon> well maybe it is maybe it isn't
[11:26] <rjharrison> Yep naughty
[11:26] <rjharrison> <header.h> functions decs
[11:26] <rjharrison> <header.cpp> the functions
[11:26] <jcoxon> which is misc/dlpipe.h
[11:27] <jcoxon> pah - just move it back into the original file then
[11:27] <rjharrison> Ok i'll have a play and get a mesaage each way
[11:27] <rjharrison> :)
[11:27] <jcoxon> misc.cpp
[11:27] <rjharrison> dlpipe.cpp
[11:27] <rjharrison> ?
[11:27] <rjharrison> need to get sandwitch
[11:28] <jcoxon> well it used to be just be in main.cpp
[11:28] <jcoxon> at the top
[11:28] <jcoxon> so put it back there i suggest
[11:45] <Randomskk> wow, django is both nicer than I remember and has some fun new features
[11:45] <rjharrison> back
[11:46] <rjharrison> django?
[11:46] <Randomskk> python web framework
[11:46] <Randomskk> like cakephp for php or rails for ruby
[11:46] <Randomskk> similar to pylons, cherrypy, web.py, webapp
[11:47] juxta (fourtytwo@ppp118-210-24-33.lns20.adl2.internode.on.net) left irc: Ping timeout: 264 seconds
[11:49] <Randomskk> especially its admin interface, that is way way nicer than before
[11:49] <Randomskk> very nifty
[11:50] juxta_ (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) joined #highaltitude.
[11:51] <rjharrison> Randomskk you keep giving very usefull info
[11:52] <rjharrison> I have been doing some php app dev and I think that cakephp may well help alot
[11:52] <Randomskk> no problem
[11:53] <Randomskk> cakephp is quite nice
[11:53] <Randomskk> though ultimately it has to use php, so you don't get quite as much magic and sugar as rails etc
[11:53] <Randomskk> rails has perhaps too much magic, some would argue, though
[11:54] <rjharrison> I want to dev some webforms apps linked to a db I guess cake is a good way forward
[11:55] <Randomskk> cake should make that kind of thing very easy
[11:55] <rjharrison> Cool :)
[11:55] <Randomskk> iirc it has a form helper and a model thing
[11:55] <Randomskk> you just make models to reflect your database and then it pretty much does everything for you
[11:55] <Randomskk> a la validation, showing errors on forms, handling authentication and security etc etc
[12:02] juxta- (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) joined #highaltitude.
[12:03] juxta_ (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) left irc: Ping timeout: 260 seconds
[12:15] <rjharrison> Randomskk I think I'm going to have to spend some time on cake
[12:15] <Randomskk> it is a fairly big thing
[12:15] <rjharrison> just been browsing
[12:15] <Randomskk> it's a fully blown MVC framework
[12:16] <Randomskk> but by and large using web frameworks makes your life much easier
[12:17] <jcoxon> bbl
[12:17] jcoxon (~jcoxon@ left irc: Quit: Leaving
[12:39] <rjharrison> Randomskk make seems to die with dl-fldigi with some crap about cs.gmo error 1
[12:39] <rjharrison> Have we seen this
[12:39] <Randomskk> I don't think so, what's the first error?
[12:40] <Randomskk> you're looking for a line from gcc starting with error:
[12:40] <rjharrison> Making all in po
[12:40] <rjharrison> make[1]: Entering directory `/home/rharrison/src/dl-fldigi/po'
[12:40] <rjharrison> test ! -f ./fldigi.pot || \
[12:40] <rjharrison> test -z "cs.gmo fr.gmo" || make cs.gmo fr.gmo
[12:41] <rjharrison> make[2]: Entering directory `/home/rharrison/src/dl-fldigi/po'
[12:41] <rjharrison> rm -f cs.gmo && : -c --statistics -o cs.gmo cs.po
[12:41] <rjharrison> mv: cannot stat `t-cs.gmo': No such file or directory
[12:41] <rjharrison> weird
[12:41] <Randomskk> that's really weird, is that just 'make'?
[12:42] <Randomskk> do you have msgfmt installed?
[12:42] <Randomskk> I see:
[12:42] <Randomskk> make[2]: Entering directory `/home/adam/Code/dl-fldigi/po'
[12:42] <Randomskk> rm -f cs.gmo && /usr/bin/msgfmt -c --statistics -o cs.gmo cs.po
[12:43] <Randomskk> and it works fine
[12:43] <rjharrison> ok one sec
[12:43] <Randomskk> seems to be part of gettext
[12:45] <rjharrison> :)
[12:45] <rjharrison> apt-get gettext :)
[12:49] <rjharrison> hum just got the count error
[12:49] <Randomskk> that should be fixed in the latest version
[12:49] <rjharrison> fetch
[12:49] <Randomskk> "git pull jamescoxon master"
[12:49] <Randomskk> you can do it with fetch
[12:49] <Randomskk> "git fetch jamescoxon"
[12:50] <Randomskk> "git merge jamescoxon/master"
[12:50] <Randomskk> the latter is the slightly more verbose way as you may want to fetch without merging or merge in some other fashion
[12:55] <sbasuita> so what's on the todo list for dl-fldigi?
[12:55] <Randomskk> some kind of web interface, a information pane showing stuff like current bearing and distance and payload info and current status and a latest message or such
[12:55] <Randomskk> actual successful extraction and uploading
[12:55] <Randomskk> the --hab mini-mode
[12:55] <Randomskk> other stuff?
[12:58] <rjharrison> http://www.robertharrison.org/mobile.php
[12:59] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 276 seconds
[13:05] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[13:10] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 258 seconds
[13:22] <rjharrison> finally a clean compile
[13:25] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[13:30] <SpeedEvil> Laurenceb: http://www.skpang.co.uk/catalog/product_info.php?products_id=660
[13:48] DanielRichman (~DanielRic@unaffiliated/danielrichman) joined #highaltitude.
[14:07] _sh3 (debian-tor@gateway/tor-sasl/sh3/x-62271040) left irc: Ping timeout: 245 seconds
[14:12] juxta- (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) left irc: Ping timeout: 276 seconds
[14:16] jcoxon (~jcoxon@ joined #highaltitude.
[14:16] <rjharrison> hi jcoxon
[14:16] <rjharrison> Just getting it all to compile under linux
[14:17] <rjharrison> do you have portaudio and pulseaudio libsnd etc ?
[14:17] <DanielRichman> rjharrison, apt-get build-dep fldigi
[14:17] <DanielRichman> and apt-get install libcurl-dev
[14:17] <DanielRichman> got everything for me
[14:17] Action: Randomskk has everything 'yes' under configure (except debug, statically linked)
[14:17] Xenion (~robert@p579FCC6D.dip.t-dialin.net) joined #highaltitude.
[14:18] <rjharrison> cool
[14:18] <rjharrison> ubuntu doesn't seem to be uptodate enough for the first to work
[14:18] <Randomskk> rjharrison: in particular then, one sec
[14:18] <rjharrison> I have compiled all the libs from source and all is happy now
[14:18] <Randomskk> hmm
[14:18] <Randomskk> it should work from apt
[14:19] <Randomskk> compiling all the libs from source is never much fun
[14:20] <sbasuita> i didn't have to compile any libs on karmic
[14:20] <Randomskk> same
[14:20] <jcoxon> rjharrison, yeah it works well on os x
[14:20] <Randomskk> I was able to install everything from apt to get it happy
[14:20] <Randomskk> sndfile, oss, portaudio, pulseaudio, hamlib, xmlrpc
[14:22] <jcoxon> sbasuita, i'll pull your fixes in a bit - nothing urgent i assume
[14:22] <sbasuita> jcoxon, yeah just some minor fixes
[14:23] <sbasuita> jcoxon, btw is the 'request pull' interface worth using or should i just ping you on irc in future?
[14:23] <Randomskk> fwiw all it does is send you a message on github
[14:23] <sbasuita> cool
[14:23] <DanielRichman> can you complete the pull through the online interface?
[14:23] <rjharrison> wish I had known about apt-get build-dep
[14:23] <Randomskk> DanielRichman: no
[14:24] <DanielRichman> hmm ok
[14:24] <jcoxon> its okay but you need to say what it is :-)
[14:24] <Randomskk> though the message it send syou includes their repo name
[14:24] <Randomskk> so you can just do "git pull <repo_url_here> master", copying the url from the message
[14:24] <jcoxon> i think you left the body empty so i didn't know if it was little fixes or a big new feature
[14:24] <Randomskk> obviously if you already had their repo as a remote with a shortname you could just "git pull shortname master" etc, assuming in both cases it's the master branch you wanted to pull
[14:24] <sbasuita> jcoxon, ah right. I assumed it would show you the commit message or something
[14:24] Xenion (robert@p579FCC6D.dip.t-dialin.net) left #highaltitude ("Verlassend").
[14:31] juxta_ (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) joined #highaltitude.
[14:40] jcoxon (~jcoxon@ left irc: Ping timeout: 264 seconds
[14:48] jcoxon (~jcoxon@ joined #highaltitude.
[15:03] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[15:24] jcoxon (~jcoxon@ left irc: Ping timeout: 240 seconds
[15:31] jcoxon (~jcoxon@ joined #highaltitude.
[15:41] wasutton3-1 (~will@ left irc: Ping timeout: 246 seconds
[15:45] <jcoxon> interestingly one of the main devs on fldigi is a uk ham
[15:48] wasutton3-1 (~will@ joined #highaltitude.
[15:49] darknesslord_ (~darknessl@ joined #highaltitude.
[16:00] LazyLeopard (~irc-clien@chocky.demon.co.uk) joined #highaltitude.
[16:10] _sh3 (debian-tor@gateway/tor-sasl/sh3/x-62271040) joined #highaltitude.
[16:22] LazyLeopard (~irc-clien@chocky.demon.co.uk) left irc: Quit: Bye
[16:39] LazyLeopard (~irc-clien@chocky.demon.co.uk) joined #highaltitude.
[16:51] Enceladus (~Brian@host86-155-6-161.range86-155.btcentralplus.com) joined #highaltitude.
[16:58] Action: Enceladus is away:
[16:59] <jcoxon> ping rjharrison
[17:17] <jcoxon> bbl
[17:17] jcoxon (~jcoxon@ left irc: Quit: Leaving
[17:31] juxta_ (~blah@ppp118-210-24-33.lns20.adl2.internode.on.net) left irc: Ping timeout: 276 seconds
[18:29] <rjharrison> boo
[18:29] <Randomskk> hi
[18:30] <rjharrison> hi Randomskk
[18:30] <rjharrison> Right read that suff on git
[18:30] <rjharrison> helpfull
[18:30] <Randomskk> there's a lot of git guides out there, github themselves have some useful ones, but that one in particular seems to help explain how git really works
[18:30] <rjharrison> and now I'm ready to have a go at sorting out the pipe and the inter process comms
[18:30] <Randomskk> the key thing is that it's essentially just a load of commits which each have an ancestor, and then branches, tags, etc etc are just pointers to commits
[18:31] <Randomskk> cool
[18:47] fsphil (~phil@2001:470:1f09:483:21f:c6ff:fe44:b25b) joined #highaltitude.
[18:48] icez (~icez@unaffiliated/icez) joined #highaltitude.
[18:53] natrium42 (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[19:04] <rjharrison> Randomskk I have nuked a file by accident can I use git to pull it back in
[19:05] <Randomskk> rjharrison: yea
[19:05] <Randomskk> "git checkout filename"
[19:05] <Randomskk> should work
[19:06] <Randomskk> natrium42: so actually django is really nice?
[19:06] <Randomskk> and sqlalchemy a bit annoying?
[19:06] <rjharrison> Randomskk cool and it worked
[19:06] <Randomskk> like, I had a play with both to try and make the same thing and ended up frustrated and annoyed at sqlalchemy but it's already done in django and now I'm just adding fancy polish
[19:07] jcoxon (~jcoxon@ joined #highaltitude.
[19:07] <rjharrison> jcoxon boo
[19:07] <natrium42> hrm
[19:07] <rjharrison> I'm on with the pipes now
[19:08] <Randomskk> natrium42: that was pylons+sqlalchemy
[19:08] <Randomskk> django is way more polished than last time I played with it
[19:08] <Randomskk> also its builtin admin interface has gone from "could be useful but mostly gets in the way" to "really useful"
[19:09] <jcoxon> hi
[19:09] <rjharrison> natrium42 you got a mention in my local rag as a Canadian Man :)
[19:09] <natrium42> rjharrison, ah cool, cheers :)
[19:10] <natrium42> yo jcoxon
[19:10] <natrium42> Randomskk, it just looks more constraining
[19:10] <rjharrison> http://www.robertharrison.org/images/various/huddersfield-examiner%2020-Mar-2010.jpg <----
[19:10] <jcoxon> hey natrium42
[19:10] <rjharrison> jcoxon i'm creatinf dl_fldigi.h
[19:11] <natrium42> Randomskk, django looks like they made all decisions for you
[19:11] <natrium42> and it seems that we will be stumbling over their constraints...
[19:11] <Randomskk> natrium42: they make some decisions along the lines of which ORM, which templater (or at least, it's got a default) but I'm not convinced those are bad decisions to hav made
[19:11] <Randomskk> in the same way that they decide on "python" for you
[19:12] <Randomskk> natrium42: there's still a lot of freedom in how you do your own application and you can use or not use parts of django as you wish
[19:12] <Randomskk> but some of the parts they offer are much better than alternatives e.g. their ORM, their templater, their view system are all great
[19:12] <natrium42> pylons doc is also better IMO
[19:13] <Randomskk> also their extras - most notably the excellent admin interface - are great and have no equivalent in other frameworks
[19:13] <Randomskk> hmm
[19:13] <Randomskk> the pylons docs are okay but http://docs.djangoproject.com/en/1.1/
[19:13] <natrium42> still for 1.1 though
[19:13] <Randomskk> those docs seem really good, from my previous and most recent experience
[19:13] <Randomskk> 1.1's also what's in e.g. ubuntu repositories etc
[19:14] <Randomskk> http://docs.djangoproject.com/en/dev/ are docs for 1.2, have docs for all new features
[19:14] <Randomskk> they release feature docs as they release features
[19:15] <Randomskk> pylons may still be the best suited for the tracker, up to you, just saying that I'm no longer convinced of what I said earlier wrt django
[19:15] <Randomskk> plus pylons is still quite nice, but while it does give you lots of freedom I'm less convinced now that that freedom is all a good thing
[19:15] <Randomskk> django've worked hard to make their platform work
[19:15] <jcoxon> rjharrison, so whats going in dl_fldigi.h?
[19:15] <natrium42> hehe, fair enough
[19:16] <natrium42> yes, the tracker is not too similar to a "newspaper site" or some content system
[19:16] <Randomskk> mainly that after toying with rails for several months I suddenly realise I like django even more than rails
[19:16] <Randomskk> yea, that's the main thing against it
[19:16] <Randomskk> it's not really a content management system
[19:16] <natrium42> which django was designed to address
[19:16] <rjharrison> jcoxon the code thats in the fork basically and some globals
[19:16] <Randomskk> but on the other hand it's not really that far off - the content is flight logs and displaying them is blitting them to a google map
[19:16] <jcoxon> in a header?
[19:16] <rjharrison> I just want minimal changes to main.cpp
[19:16] <Randomskk> with fancy searching and selecting of content
[19:17] <rjharrison> #include <dl_fldigi.h>
[19:17] <Randomskk> plus the backend stuff which mostly amounts to adding new content
[19:17] <Randomskk> and selecting specific bits and pieces of content
[19:17] <jcoxon> isn't that what i did with dlpipe.h
[19:17] <Randomskk> almost like a multi-user blog or such
[19:17] <rjharrison> jcoxon, the cose will be in dl_fldigi.cpp
[19:17] <rjharrison> code
[19:17] <jcoxon> oh right
[19:17] <jcoxon> okay
[19:17] <jcoxon> remember to 'add' it to git
[19:18] <rjharrison> :) Yep
[19:18] MoALTz_ (~heh@ joined #highaltitude.
[19:18] <rjharrison> Basically header for global vars (pahh) we should be passing the vars through but that's alot of work
[19:21] MoALTz (~heh@ left irc: Ping timeout: 252 seconds
[19:22] <jcoxon> okay
[19:25] <jcoxon> sounds good
[19:25] <jcoxon> ive done some gui stuff
[19:26] <jcoxon> but its only the beginning
[19:26] <sbasuita> of the end
[19:26] <sbasuita> *cue dramatic music*
[19:28] <DanielRichman> *cut to meeting room with 1 empty chair* "He was the best tracker we had"
[19:28] <DanielRichman> *cut again, shot of payload falling 100m short of the beach*
[19:29] <fsphil> ah great, now I've got the baywatch theme in my head
[19:29] Action: SpeedEvil runs slowly across the channel.
[19:30] <Randomskk> oh witty
[19:31] <fsphil> "I'll be there... "
[19:33] <jcoxon> DanielRichman, i put some code in the config so now you can turn your wfall to png function on and off
[19:34] <fsphil> what's the idea behind the web interface?
[19:34] <DanielRichman> jcoxon, sorry, do you mean you've done that or need me to add that to the wfall png code?
[19:37] <DanielRichman> Ah, it looks like you've done it
[19:39] Enceladus (~Brian@host86-155-6-161.range86-155.btcentralplus.com) left irc: Quit: AmigaOS 4.1. Remember when computing was fun?
[19:40] <natrium42> hehe, rock on
[19:43] <natrium42> @AmigaOS
[19:43] <fsphil> Been a while :)
[19:45] <fsphil> I still have my A600, working perfectly. got an amazing 1MB RAM upgrade to 2MB
[19:45] <natrium42> :D
[19:46] <Laurenceb> SpeedEvil: nice, very cheap
[19:46] <Laurenceb> but not as capable as the Si4432 with the proper matching
[19:47] Simon-MPFH (~simon@phantom.mpfh.co.uk) left irc: Quit: Leaving
[19:47] <SpeedEvil> Laurenceb: no. Small, cheap, lightweight tho
[19:47] <SpeedEvil> approved
[19:48] <Laurenceb> just looking at the datasheet
[19:48] <Laurenceb> its no a bare die 4432
[19:48] <Laurenceb> the sensitivty is lower
[19:48] <Laurenceb> -116 dbm for 1.2kbps, the 4432 is -122
[19:51] <SpeedEvil> odd
[19:51] <SpeedEvil> seconds?
[19:56] <rjharrison> humm automake is over rated
[19:56] <rjharrison> well I can't make it do what I want
[19:57] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Read error: Connection reset by peer
[19:58] <Randomskk> autotools generally is so big and painful, I hear
[19:58] Action: Randomskk has never really used it for anything serious
[19:58] <fsphil> heh, I must be getting soft in my old age. I quite like C++ now :)
[19:58] <Randomskk> I hear "autotools hell" a lot
[19:58] <Randomskk> C++ is pretty nice really
[19:59] <Randomskk> also it's a "serious" language
[19:59] <fsphil> I tended to stick with C
[19:59] <Randomskk> I'll still use C on microcontrollers
[19:59] <Randomskk> but for a computer, C++ is a bit nicer
[19:59] <Randomskk> OO is generally a good idea, I feel
[20:00] <natrium42> you can do OO in C just fine :P
[20:00] <natrium42> sans the ugly syntaz
[20:00] <natrium42> -z+x
[20:00] <rjharrison> I'm trying add a new misc/dl_fldigi.cpp to the Makefile.am
[20:00] <Randomskk> ugly syntax and generally how it was never designed for it and thus you just have to abuse structs? :P
[20:00] <Randomskk> you can't have object methods really
[20:00] <Randomskk> well
[20:00] <Randomskk> it's still icky
[20:00] <natrium42> function pointers
[20:00] <Randomskk> yea
[20:00] <Randomskk> but it's not really the same
[20:00] <natrium42> and a bunch of macros
[20:01] <natrium42> meh
[20:01] <Randomskk> scoping for one thing
[20:01] <sbasuita> you could do it in c, but really you'd just spend your time reinventing c++
[20:01] <natrium42> i wish the C++ weren't so afraid to add new keywords
[20:01] <fsphil> rjharrison, add it and then run autoreconf -vfi
[20:01] <natrium42> *the C++ guys
[20:02] <Randomskk> excellent, all the text stuff for my application is totally done already, that was painless
[20:02] <Randomskk> now to add pretty graphs
[20:02] <Randomskk> and then perhaps consider some kind of user system
[20:03] <Randomskk> natrium42: django also has a pretty nice authentication system :P
[20:03] <natrium42> hrm
[20:03] <natrium42> fine i will look at it again :P
[20:03] <Randomskk> woo, mission success
[20:03] <natrium42> Randomskk, what are you working on?
[20:04] <Randomskk> natrium42: right now something to scrape the price of some mutual funds from fidelity, archive, graph, maths
[20:04] <Randomskk> I would just be using google finance, but they don't index all the mutual funds in my isa
[20:04] <natrium42> neat
[20:04] <Randomskk> the scraping is actually silly easy
[20:04] <Randomskk> it's two lines
[20:04] <Randomskk> well three I guess
[20:05] <Randomskk> page = urllib2.urlopen(self.url)
[20:05] <Randomskk> soup = BeautifulSoup.BeautifulSoup(page)
[20:05] <Randomskk> return float(soup.find("span", id="buyprice").string)
[20:06] <Randomskk> so I have a fund model, a fundprice model that is used to store date-price, an isa model that has many funds through a relationship that stores how many of them and how much they were bought for
[20:06] <Randomskk> then some basic methods on the models can return things like price, profit, value etc
[20:06] <Randomskk> shove all that together with some simple templates and the whole system is go
[20:07] <Randomskk> I can add new funds, new funds to isas and new isas from the django admin interface really easily, since it sets up all the related models in the right forms
[20:07] <fsphil> that sounds awesome
[20:07] <Randomskk> so the admin is entirely done for me and that includes user authentication
[20:08] <Randomskk> all I've programmed is what data the models have and some methods for things like "profit = self.latest_value() - self.initial_value()" etc, the templates take those values and I'm set.
[20:10] <rjharrison> `:q
[20:10] <Randomskk> and the admin interface supports things like filtering, searching, grouping by date etc by just adding the appropriate words
[20:10] <rjharrison> opps :)
[20:10] <fsphil> hehe
[20:11] <Randomskk> https://randomskk.net/u/isa_admin.png etc
[20:12] <Randomskk> and then in my app https://randomskk.net/u/funds.png etc (check out the eerily similar colour scheme to http://cuspaceflight.co.uk/calc :P)
[20:13] Action: Randomskk goes back to playing with graphs
[20:34] <natrium42> Randomskk, nice, what did you use for the widgets?
[20:35] <Randomskk> natrium42: nothing?
[20:35] <Randomskk> natrium42: the ones on the admin page are the default normal admin things
[20:35] <natrium42> in django?
[20:35] <Randomskk> yea
[20:36] <Randomskk> you just
[20:36] <Randomskk> from models import ISA
[20:36] <Randomskk> admin.site.register(ISA)
[20:36] <Randomskk> and then the built in admin thing will automatically include it
[20:37] <Randomskk> so does stuff like add, delete, edit etc
[20:38] <natrium42> sounds convenient
[20:39] <Randomskk> it is
[20:40] <Randomskk> http://docs.djangoproject.com/en/1.1/intro/tutorial02/ has details on using it with various models
[20:41] <Randomskk> e.g. telling it what fields to use, what related models to add in and how to display them (in a table vs a stack etc)
[20:43] jcoxon (~jcoxon@ left irc: Read error: Connection reset by peer
[20:48] <SpeedEvil> http://www.dealextreme.com/details.dx/sku.34396
[20:48] <SpeedEvil> polypropylene vacuum flasks?
[20:57] jcoxon (~jcoxon@ joined #highaltitude.
[20:57] <jcoxon> hey rjharrison, making progress?
[20:58] <DanielRichman> Internet Explorer, I hate you yet again
[20:59] <natrium42> http://ripie6.com
[20:59] <DanielRichman> oh no, I hate IE7 and IE8
[20:59] <DanielRichman> All I wanted to do was serve up application/xhtml+xml
[20:59] <DanielRichman> but no.
[20:59] <natrium42> hehe
[21:09] <rjharrison> jcoxon yep
[21:09] <rjharrison> nearly there
[21:09] <rjharrison> ready for you to integrate the code
[21:11] <rjharrison> jcoxon I'm ready for you to pull the changes from my repo
[21:12] <Laurenceb> http://imagebin.org/89810 <- routing is mostly done
[21:12] <Laurenceb> needs tydying up and ground planes
[21:14] <rjharrison> jcoxon
[21:16] <SpeedEvil> http://www.vimeo.com/8600426 insane - helicopter as you've never seen it before
[21:18] <rjharrison> ping jcoxon
[21:23] <Laurenceb> bbl
[21:28] <rjharrison> jcoxon just realised I hadn't git add the 2 files
[21:28] <rjharrison> done that and pushed
[21:33] <rjharrison> bbl
[21:36] <jcoxon> yeah yeah yeah
[21:37] <jcoxon> rjharrison, is it 2 way now then?
[21:39] <earthshine> Evening
[21:39] <jcoxon> hey earthshine , how was fishing?
[21:39] <earthshine> Dissapointing
[21:40] <earthshine> We caugh plenty of fish - but everyone was a codling that had to go back
[21:42] <rjharrison> back
[21:43] <rjharrison> jcoxon no it's just implemented nicely now
[21:43] <rjharrison> ie dl_fldigi.h and dl_fldigi.cxx
[21:43] <jcoxon> okay cool
[21:44] <jcoxon> just compiled it and pushed it onto github
[21:44] <rjharrison> you will need to autoreconf again I'm afraid
[21:44] <jcoxon> done
[21:44] <jcoxon> so did you open fldigi with the --hab cmd arg?
[21:45] <rjharrison> jcoxon yep I like it
[21:45] <rjharrison> there are some screen draw issues with it on linux
[21:45] <rjharrison> But the idea of a reduced window is cool
[21:46] <jcoxon> yeah i'm getting to the bottom of that
[21:46] <jcoxon> but will need a linux box to really sort it out - currently it works okay on my mac
[21:47] <rjharrison> Cool
[21:47] <rjharrison> I could send you an eee
[21:47] <jcoxon> also did you see DanielRichman's png hack
[21:47] Enceladus (~Brian@host86-155-6-161.range86-155.btcentralplus.com) joined #highaltitude.
[21:47] <rjharrison> jcoxon yep
[21:47] <jcoxon> hehe i've got a linux box at home
[21:47] <rjharrison> That looks cool for your html hack
[21:47] <jcoxon> going to persuade someone, (Randomskk perhaps) to make a proper interface...
[21:48] <rjharrison> Randomskk seems to be a good coder
[21:48] Sophira (~sophie@neo.theblob.org) left irc: Quit: Back in a bit, probably; reorganising my windows.
[21:48] <rjharrison> better than me anyhow and I'm a touch busy these days
[21:49] <rjharrison> I like the idea of opening another window for curl results
[21:49] <rjharrison> and I'll post those back to the parent
[21:50] <jcoxon> as in gui window?
[21:50] <rjharrison> Yep
[21:50] <jcoxon> yeah
[21:50] <rjharrison> I also think current lat long should be posted with the data for when we're gps enabled
[21:51] <rjharrison> callsign,lat,lon,data ....
[21:51] <jcoxon> i was thinking we could add a hab bar to the --hab bit, with decoded data and server replies
[21:51] <rjharrison> Yep that would be cool too
[21:52] <jcoxon> i'll see what i can do
[21:52] <rjharrison> I'm beginning to like the git
[21:52] <rjharrison> Bit of a bigger learning curve than svn
[21:53] <jcoxon> oh yes - its more complicate then svn
[21:54] <jcoxon> i'm finding it hard not to fall completely in the trap of just porting all the old code over
[21:54] Enceladus (~Brian@host86-155-6-161.range86-155.btcentralplus.com) left irc: Quit: AmigaOS 4.1. Remember when computing was fun?
[21:55] <rjharrison> jcoxon hehe
[21:56] <rjharrison> did you see how I have been tagging dl code changes
[21:56] <rjharrison> I have moved the vars to dl_fldigi... too
[21:56] <rjharrison> rather than rjh etc ...
[21:56] <jcoxon> oh i see
[21:57] <jcoxon> i like doing //jcoxon as i can easily search for it within the code
[21:57] <rjharrison> as dl_fldigi is so much harder :P
[21:57] <rjharrison> :)
[21:57] Enceladus (~Wookiecha@host86-155-6-161.range86-155.btcentralplus.com) joined #highaltitude.
[21:57] <jcoxon> really we don't need to do it at all as git can find it all for us anyway
[21:58] <rjharrison> true
[21:58] <rjharrison> It's pretty cool
[21:58] <jcoxon> i like to know what i've done so i'lll stick to //jcoxon - its pretty harmless
[21:58] <rjharrison> yep would you like us to adopt to our names so you know where to shout later?]
[21:59] <jcoxon> its okay
[21:59] <Randomskk> check out `git blame`
[21:59] <jcoxon> the problems are usually my own making
[21:59] <jcoxon> :-p
[21:59] <rjharrison> I'm interested to know when we will deal with the input from the child to the parent
[21:59] <rjharrison> I guess it will be in that inner loop you must be in already
[22:00] <jcoxon> yeah, trx.cpp
[22:00] <jcoxon> if we perhaps have a buffer which gets checked and moved to a nice window
[22:01] SpeedEvil1 (~user@tor/regular/SpeedEvil) joined #highaltitude.
[22:01] <rjharrison> yep
[22:01] <rjharrison> f**k me git blame is cool
[22:01] <rjharrison> try git blame src/main.cxx
[22:03] <Randomskk> http://github.com/jamescoxon/dl-fldigi/blame/master/src/main.cxx
[22:03] <DanielRichman> jcoxon stole all my git blamez when he reindented everything :D
[22:03] <Randomskk> github has a really nice blame interface
[22:03] <DanielRichman> rjharrison, you can svn blame too
[22:03] <sbasuita> you can also bzr praise if you're feeling like less of a git ;P
[22:03] <rjharrison> DanielRichman cool
[22:03] <rjharrison> Never even knew it was there
[22:04] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 252 seconds
[22:04] <rjharrison> jcoxon gets the blame :)
[22:04] <jcoxon> pah
[22:04] <rjharrison> right bed beckons but I think were on the way now
[22:05] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Ping timeout: 276 seconds
[22:05] <jcoxon> all that blame - i just do the hard work and make the tough decisions!
[22:05] Nick change: SpeedEvil1 -> SpeedEvil
[22:05] <rjharrison> The posting via curl should still work atm I guess it's just the config that needs to be done :)
[22:05] <jcoxon> yeah it works - i tested it with default values and a wav file
[22:05] <jcoxon> will get the config done at some point
[22:05] <rjharrison> jcoxon do you concur?
[22:06] <rjharrison> cool
[22:06] <rjharrison> right bed
[22:06] <jcoxon> night
[22:11] <Enceladus> Goodnight all
[22:11] Enceladus (~Wookiecha@host86-155-6-161.range86-155.btcentralplus.com) left irc: Quit: AmigaSYS 4 http://amigasys.extra.hu
[22:12] jcoxon (~jcoxon@ left irc: Quit: Leaving
[22:16] DanielRichman (~DanielRic@unaffiliated/danielrichman) left irc: Quit: Leaving
[22:34] darknesslord_ (~darknessl@ left irc: Quit: darknesslord_
[22:35] darknesslord_ (~darknessl@ joined #highaltitude.
[22:45] DanielRichman (~DanielRic@unaffiliated/danielrichman) joined #highaltitude.
[22:49] DanielRichman (~DanielRic@unaffiliated/danielrichman) left irc: Remote host closed the connection
[22:56] fsphil (~phil@2001:470:1f09:483:21f:c6ff:fe44:b25b) left irc: Quit: Brainsszzzz.....
[23:03] Hiena (~Hiena@ left irc: Quit: -=Got bored from the net. Gone blowing up things.=-
[23:07] Laurenceb (~laurence@host81-154-84-43.range81-154.btcentralplus.com) left irc: Ping timeout: 248 seconds
[23:16] Jasperw (~jasperw@client-86-25-212-88.cht-bng-015.adsl.virginmedia.net) left irc: Quit: Leaving.
[23:17] preetum (~apollo299@c-67-160-116-121.hsd1.wa.comcast.net) joined #highaltitude.
[23:17] <preetum> KE7SRY here
[23:18] <preetum> I'm located in Redmond, WA (USA), and I'm trying to get in to ballooning.
[23:18] <Randomskk> hi preetum
[23:21] <Randomskk> a lot of the people normally around this channel are in the UK so afk
[23:21] <preetum> hey, Randomskk. I have experience with mobile/embedded development, so the camera+payload shouldn't be a problem. Right now, I'm trying to figure out air<->ground communications.
[23:21] <Randomskk> mostly what we do is a 10mW 434MHz transmitter on the payload sending RTTY at 50 baud
[23:21] <Randomskk> receive with a standard ham radio capable of 434MHz SSB and a yagi antenna
[23:21] <Randomskk> we then have computer software that will decode it quite well
[23:21] <Randomskk> you don't get much data but enough for GPS
[23:21] <preetum> hmm. the other thing is: I probably won't be able to recover it, so I'm trying to transmit low-grade pictures from the balloon directly.
[23:21] <preetum> (as opposed to extracting the pictures upon recovery)
[23:22] <Randomskk> difficult but has been done and someone (fsphil) is working on making it better
[23:22] <Randomskk> SSTV has been used to some success
[23:22] <Randomskk> I would aim to recover though
[23:22] <Randomskk> have you seen the flight path predictor?
[23:22] <Randomskk> it's quite accurate and good at letting you know if it's likely to end up in the sea or not
[23:23] <Randomskk> http://cuspaceflight.co.uk/calc
[23:23] <preetum> yea, it's predicted to land in the midst of a national park, 7 hrs-ish away
[23:23] <Randomskk> bit of a pain, but the weather changes quite dramatically
[23:23] <Randomskk> you may find it ends up quite close soon
[23:23] <preetum> true. I won't leave recovery out of the question.
[23:24] <preetum> but I'd like to have a "backup" plan to retrieve the pictures
[23:24] <Randomskk> but yea, it's definitely possible to send images - either SSTV or by sending actual image data
[23:24] <Randomskk> just less reliable and you're only going to get low resolution images
[23:24] <SpeedEvil> why probably won't be able to recover it?
[23:25] <preetum> SpeedEvil: because the predicted landing site is in a relativly inaccesable part of a national park
[23:25] <preetum> ie, forest
[23:25] <SpeedEvil> Ah.
[23:25] <SpeedEvil> Get your hiking boots!
[23:25] <Randomskk> preetum: it could well change quite dramatically
[23:25] <Randomskk> also that would be a fun hike
[23:25] <Randomskk> :P
[23:25] <preetum> hehe
[23:25] <preetum> perhaps I'll go with some friends, as a weekend excursion
[23:25] <SpeedEvil> predicted landings are - typically - no better than +-30km or so
[23:25] Action: Randomskk would totally go hiking after a payload
[23:26] <SpeedEvil> and that's for a prediction made on the day of launch
[23:26] <russss> any idea what the law is on steerable return vehicles in the US? ;)
[23:26] <SpeedEvil> (IIRC - when I looked at some results)
[23:26] <preetum> russss: being a model aircraft enthusiast, I actually considered that for a bit...
[23:27] <russss> preetum: we're fairly sure it's illegal in the UK but I think there are some people who are toying with what constitutes a "parachute".
[23:27] <Randomskk> the law is very grey unfortunately
[23:28] <preetum> even if steerable "parachutes" are legal, I don't think onboard batteries would be able to overcome winds and such
[23:28] <preetum> though I suppose it depends on how heavy the payload can be.
[23:29] <Randomskk> a microsd card and a glider
[23:29] <Randomskk> ride with the winds
[23:29] <russss> you've got a lot of altitude to play with
[23:29] <preetum> but if the winds push a balloon off course, won't the "glider" need to fight against the winds?
[23:29] <russss> I think you could get a couple of 10s of km
[23:29] LazyLeopard (~irc-clien@chocky.demon.co.uk) left irc: Quit: Bye
[23:30] <preetum> I'm not really a glider guy (mostly gas/electric), so I wouldn't know the intricaties of the jet streams
[23:31] <preetum> but my intuition is telling me that the craft would need to combat winds to return to its original location
[23:31] <Randomskk> I think the main thing is that you have so much altitude and GPE to play with
[23:31] <Randomskk> you would have to combat winds to an extent
[23:32] <Randomskk> but I think if you were glinding the winds are going to have much less of an effect
[23:32] <SpeedEvil> If you have comms - you can tell it 'go south' - which you know to be downwind
[23:32] darknesslord_ (~darknessl@ left irc: Quit: darknesslord_
[23:32] <russss> a powered vehicle of any size would definitely not be allowed in the UK ;)
[23:33] <preetum> I think I'll try to first get something basic up there, then maybe try something more adventurous on my second trip.
[23:34] <preetum> launch*
[23:34] <preetum> about SSB: I've located a 10mW transmitter. Something like this? - http://www.lemosint.com/product_Details.php?itemID=206
[23:34] <preetum> one of the UK wikis referenced it.
[23:34] <russss> that's the one we use yeah
[23:35] <SpeedEvil> why not just wait until you get a launch window that goes to a easy to retrieve spot?
[23:35] <russss> assuming that band is license exempt in the US...
[23:35] <SpeedEvil> Or do you have geograpy problems?
[23:36] <preetum> SpeedEvil: we live in a fairly populous area, so I don't want the balloon to come down there.
[23:36] <preetum> the winds push the balloon toward Canada.
[23:37] <preetum> that transmitter says 400Mhz... which means it won't work with my VHF handheld right?
[23:38] <russss> preetum: I think the larger issue is that it would be illegal to transmit in the band in the US
[23:38] <russss> "Available for licence-exempt operation in the 433MHz (EU) and 458MHz (UK) bands"
[23:40] <preetum> I think I can operate in those bands, since I have a Tech. ham license
[23:40] <preetum> err, I may need to upgrade to General
[23:41] <preetum> http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&rgn=div6&view=text&node=47:
[23:42] wasutton3-1 (~will@ left irc: Ping timeout: 240 seconds
[23:43] <preetum> never mind, a Tech. class will suffice. I'm allowed to transmit within "50 km of the Earth's surface"
[23:43] <Randomskk> lucky
[23:44] <Randomskk> our extra-equivilent doesn't allow operating airborne
[23:46] <preetum> do I need to buy a UHF reciever in order to pick up on the signal? or is there something I can do with my VHF portable?
[23:46] <Randomskk> your vhf portable may be able to pick up 434
[23:47] <Randomskk> but ideally you'll want something you can plug a yagi into
[23:47] <Randomskk> and plug your computer into
[23:47] wasutton3-1 (~will@ joined #highaltitude.
[23:49] <preetum> okay. do you recommend any model in particular? preferable it would be something cheaper; I'm trying to keep this project on a budget
[23:50] <Randomskk> the yaesu ft817 is the model of choice for a lot of us but it's not exactly cheap
[23:51] <preetum> yea, that's a bit too high for my wallet. I don't need multiband or any fancy features.
[23:55] <Randomskk> check the specs for your handheld
[23:55] <Randomskk> it may be okay
[23:55] <preetum> I currently have an IC-V82. the U82 version of this handheld supports the required range, and runs at ~$200.
[23:56] <Randomskk> do you have any other radio? APRS gear?
[23:57] <preetum> no. I just recently got into ham radio, and haven't had much time to pursue it very much.
[23:57] <preetum> the radio needs to support SSB, right?
[23:57] <Randomskk> for our transmission style yess
[23:59] <SpeedEvil> the radio needs to support SSB - or have some coding done on recieving low bitrate signals on FM
[23:59] <Randomskk> but you could transmit however you can receive technically
[23:59] <SpeedEvil> it should in principle be possible - you will suffer more interference
[00:00] --- Mon Mar 22 2010