#indiewebcamp 2012-11-10

2012-11-10 UTC
dascher, borior, zztr and spinnerin joined the channel
#
@sandeepshetty
Stumbled upon [Dopplr colors](http://blog.dopplr.com/2007/10/23/in-rainbows/) via @briansuda. Using it in #converspace to [...](http://www.sandeep.io/posts/15) #indieweb
josephboyle, donpdonp, morrocco_mole, dietrich, aaronpk, joshr_, hober, singpolyma, lmorchard, catsup, wajiii, reidab, heathjs, brianloveswords, mkowens, elf-pavlik, jancborchardt, paroneayea and sdboyer joined the channel
#
@ocodia
Day One is a beautiful app. I love that it syncs to iCloud or Dropbox. #OwnYourData https://itunes.apple.com/gb/app/day-one-journal/id421706526?mt=8
dascher, barnabywalters and sandeepshetty joined the channel
#
barnabywalters
hey sandeepshetty — loving the auto colouring of tags on sandeep.io!
#
barnabywalters
cool. I hadn’t realised those were auto generated before
#
sandeepshetty
I like what Brian has done with it...
#
barnabywalters
I suspect I shall soon be doing some design improvements based on sandeep.io and the new instagram profiles
#
barnabywalters
lots of stuff I like in both of them
#
sandeepshetty
me neither till I discovered his blog post about dopplr colors and it struck me that they might be auto-generated so I checked a few of the color codes and they were based on dopplr color algo
#
barnabywalters
did you decide how you’re doing pagination yet?
#
sandeepshetty
Do you mean infinite scroll vs pagination or backend implementation?
#
barnabywalters
sandeepshetty: URLs ;)
#
sandeepshetty
BTW this is what it looks like to me after I login: http://imgur.com/nnhtT
#
sandeepshetty
Haven't thought about the pagination URL... might go with ';'
#
barnabywalters
not a query string?
#
sandeepshetty
and I'm leaning towards infinite scroll
#
barnabywalters
sandeepshetty: sure, but you still need URLs in there somewhere ;)
#
sandeepshetty
yeah... query string is my second option... haven't thought this through really... based on my next action list I might get to it in 48 hrs :)
#
barnabywalters
at the moment I have ?day= and ?max_day= params, which are nice and permanent, but a bit messy
#
barnabywalters
it results in a lot of empty pages if I havent posted for a while
#
barnabywalters
so if max_day/date/whatever is supplied without a `day`, I’ll do normal pagination, returning the last 20 or so items before that date
#
barnabywalters
still permanent (not like pageid=2, for example) and better experience. also easier to do infinite scrol
#
sandeepshetty
Think I remember seeing aaronpk doing it based on dates... kinda like posts older than this datetime....
#
sandeepshetty
I liked that approach
#
barnabywalters
yep, that’s how I should have done it all along — and will be doing it
#
sandeepshetty
barnabywalters - Curious to know what you might be picking up from sandeep.io & instagram profiles
#
barnabywalters
sandeepshetty: the clean look and colourful tags from sandeep.io, and the autorotating photo header from the instagram profiles
#
sandeepshetty
I need to figure out how to solve the photo problem... haven't decided which way to go.. use openphoto or build my own
#
barnabywalters
sandeepshetty: for the record, I use openphoto and am happy with it. It’s API is solid enough for me to integrate it with my own software. I might roll my own one day, but have no reason to for the moment
#
barnabywalters
for example, the rotating photo banner will be really easy to do with the openphoto API
#
sandeepshetty
So you have it installed on your own server and saving to photos on your server (not s3)?
#
barnabywalters
as would (for example) making it find photos tagged with the same tags as the items on the homepage
#
barnabywalters
yep, I have it saving everything locally
#
barnabywalters
I run a periodic backup and sql dump to my local machine and external harddrive
#
sandeepshetty
Does it save image to db or disk?
#
sandeepshetty
s/image/images
#
Loqi
sandeepshetty meant to say: Does it save images to db or disk?
#
barnabywalters
sandeepshetty: it saves the images to disk, metadata to DB
#
sandeepshetty
I might give it a spin in a few days...
#
barnabywalters
I only just realised that dropbox can be installed and managed via command line on servers, so what I might do is use that as an intermediate backup
#
barnabywalters
so openphoto would save to disk, and DB would sync it’s files between computers and back them up
#
sandeepshetty
Sounds like a good plan.
barnabywalters, dascher and sandeepshetty joined the channel
#
tommorris
barnabywalters: I'm in the middle of writing my own image storing thing. So tommorris.org/images/ will exist, and the images stored on Amazon S3. Been my excuse to learn how to use ImageMagick/GraphicsMagick, the AWS::S3 library in Ruby.
#
tommorris
plus each image will be CC licensable.
#
tommorris
likes the whole pay-as-you-go hosting thing compared to having to give Flickr money each year.
#
barnabywalters
tommorris: oh cool
#
tommorris
I store most of my images on Wikimedia Commons anyway
#
tommorris
but the ones that don't go on there will go on /images/
#
tommorris
but the /images/ thing has been put on hold. this evening, I'm working on merging Food Standards Agency data into OpenStreetMap
#
barnabywalters
tommorris: nice, what sort of data is it?
#
tommorris
just basic business metadata. street address, postcode, some unique identifiers (inverse functional properties), 'BusinessType' (which allows you to work out if it is a restaurant, take-away, church/community group etc.)
#
tommorris
I'm just using it to increase the quality of data in OSM anyway.
barnabywalters, barnabywalters_, dascher, dascher_ and lmorchard joined the channel