[click to enlarge]
–>BTW: app UPD: with turning off Random & switching to latest RSS and selecting Tiffany – You can now follow latest ImgGrapher & fiddle pics!
Finally (thanks to one of our readers!) we did release the iPad app!
2readers: Plz put a rating! This will help a lot (reviews are counted in some monitors showing new/price drop apps). I know there are a lot of readers here, if You like the site just download it on your (or friend’s 8) iPad!
Just search “Kpop iDol” in App-store!
It is both Guess Who Game – To test your Idols knowledge plus a full screen (+zoom+swipes) pic browser. You can choose from a lot of Idols (we can add if someone is missed).
P.s. Don’t forget to swipe, double-tap, edit Idols list and change random mode on-off (in settings) !
And to celebrate it:
Just download it and put 5* in review – gaining both [ViP] status & more gifts (like jps invites, etc.)!
I’m sure You’ll find it useful!
Also any feedbacks and ideas are welcomed! (GIFTs included 8)
Nice work! will be release in the future for the iPhone? it would be interesting to try it, thanks.
Ok, its really good to see the progress of the page, thanks for the information 😀
looks like tiny-g dohee
after school jooyeon
if no one answered, then i will
it’s hwayoung
Bekah and Uee
yeah, i was shocked when i read the news, since uee is my bias. i don’t wanna her to graduate soon, atleast for next AS comeback..
Hey it’s pretty good! Just downloaded it and rated it 5 in iTunes. This is not the US store btw but I guess all the reviews and downloads count!
Is is true one can get Vip? How does that work,should I capture my iTunes screen with the app downloaded? Thanks!
I liked the app its fun to spend some time looking at pretty pictures.
The look is fine but I have some suggestions. Selecting the idol then pressing ok in unnatural in an app. I feel it should be just selecting the idol and then you get you’re correct,not correct etc. Just one step, make it faster. Hints could be given when you select the wrong idol,automatically, up to two times,then it just shows : you’re wrong.
It would be nice if it would give you some kind of score or we could have some sort of “leaderboard” to compete between us readers.
Also maybe when you get it right, some links to latest youtube videos of that idol/group?
Just some suggestions ,but I really like it!
I think zooming should be with “pinch” like other apps do it. You can have both ways but when you see a picture in mobile you pinch with both finfers to zoom.
Also I think an RSS picture viewer (where you dont have to guess) its a nice idea. Who doesnt want to spend sime time looking at pretty girls. It can even be random but you need to have a lot of sources so it never “runs out of pictures”. It would be like flipping an endless magazine. You could have “ads” somewhere that are actually this site latest posts or news.
Just some ideas!
pinch – это щипок, так называемый.
используется для увеличения и уменьшения картинки на сенсорных экранах двумя пальцами.
а swipe – это скользящее движение пальцем по экрану, например, для разблокировки экрана или перелистывания.
это так, для справки 8)
по отражению сложнее определить кто это 🙂
немного похожа на Ывон из Jewelry, но видимо это не она.
есть немного что-то от девченок из APink.
проголодалась видать 8)
забавная фотка.
Это она от избытка чувств. 🙂
А фотка замечательная, да.
Let’s make a more detailed specification of what we will do 🙂
There are a few points that need clarification. Namely the following:
History
> Live data or historical data?
> Store all pictures starting from now => need server
> Need to add new images to database periodically => Use Cron job?
> Or just fetch live data from RSS feeds/twitter in live => no need for server
Actions
> Manage users???
> What does it mean???
> Users can register???
What user actions need to be analyzed?
> Track users?
> Count picture clicks?
> Top viewed pictures?
> Picture rating?
> Top rated pictures?
Subscriptions?
> What does it mean???
Sources
> Where do we get the images from?
> RSS feeds?
> Google image search?
> Sites without RSS feeds?
> Twitter?
> Default sources?
> Custom sources?
> How do custom sources deal with historical data?
> Custom sources only get live data (e.g. what’s in the RSS feed right now?)
Storage
> What needs to be stored?
> Image URL?
> Image metadata (date, tags)
> Image itself?
Targeted platforms
> Mobile devices only?
> iOS?
> Android?
> Others?
Ok it’s way more detailed now 🙂
We need to organize this a little. Making a repository on bitbucket and add issues for each task seems like a good idea.
Like write everything that needs to be done, with priorities and start with small things and see how it goes.
So to recap:
– Server that has db to store image URLs with metadata
– Update script that collects new images from different sources and stores them
– Cron job that calls update script every X minutes
– Start with small things: no custom sources, no registration, add them later when simple and most used features work
>Other questions
Name of app?
Related to 4sashi.com in some way?
2 ways to do it, which one?:
a)
– server offers web service and serves image urls in json
– client app makes requests to web service, creates view with json data
b)
– server serves view
– client just displays view
Possible solutions to mentioned problems
Sites without RSS feed
> Use HTML parser like JSoup in Java, Beautiful Soup in Python
Performance
> Python should not be a problem, unless we have a lot of users, but I think we should be ok judging by download counts I get
Client app
> Could be done entirely in HTML/JS
> Interesting libraries and frameworks:
– hammer.js (for pinch to zoom, pan, double tap, swipe…)
– angular-gestures (module for AngularJS, uses hammer.js)
– AngularJS (really good!)
– Polymer (for nice UI)
Hosting
> Google app engine?
> Python
> Cron jobs
>>Maybe to create new db on each update and next (day?) transfer it to the “main” db(?)
Just make one database and add rows only if the content is not already in it => check that the content is actually new
Yes, the “plugin” idea is what I did for the Android app.
About twitter sources, some japanese idols post pics on twitter, maybe we can use such twitter profiles to get interesting pictures
Google app engine offers cron jobs
https://developers.google.com/appengine/docs/python/config/cron
I thought it was gonna be related to 4sashi.com somehow.
Sometimes it’s best when things don’t get too popular…
For the 2 ways thing:
Either version a)
– Client makes a request to the server, which replies with JSON data. Client uses that data to build a view and displays it.
Or version b)
– Server creates a view (HTML page, already formatted) and client just displays it (nearly no work on client).
I used AngluarJS and really liked it. It was a desktop app though, didn’t try it in a mobile webview. There’s also angularUI in case we need some UI elements (maybe even better than polymer because from what I’ve seen so far, polymer looks nice but is really slow to load unfortunately and even slower on mobile devices)
hammer.js: didn’t try but since we can use it the “AngularJS way” with angular-gestures I thought it could be what we need.
I deployed a Django app on Apache httpd but for this project I think Django is overkill, something more lightweight like bottle or flask would be enough. Only way I see Django would be useful is if we plan to add more complex stuff in the future, then probably it’s best to start with Django but if we just need to serve some JSON data and have 2-3 URLs the others are enough.
Since I don’t have an iDevice I’m gonna use an Android phone to develop. Yes, the idea is to have very little native OS-specific code (just to setup a webview) and run the HTML/JS app in that webview. There is a project called Cordova which could help with writing a cross platform app. Apparently it also has bindings to access native features like camera, sensors, etc. from Javascript. Maybe we should look into this.
I think Google app engine should be good for a start, there’s not so many readers and if we need to scale then we can always upgrade from “free” google app engine to “paid” google app engine or host it somewhere else.
Though, I’m doing this mostly because I like to develop things in my free time and it gives me an opportunity to get some experience… not seeking to make money with it. I’d appreciate if we can keep it ad-free since I don’t like seeing ads myself when I use an app or browse a website.
Wow you’re right about the quota, just read the pricing page and apparently it would cost 0.06$ for 100’000 writes. It’s gonna cost a lot and don’t really like the idea of pay per writes. Indeed we need to find another host.
I did that after you told me about it last year (OSX on vmware + install xcode and use ios simulator), it worked but to be honest I prefer to test apps on a real device. A friend let me borrow his iPhone but then I had to pay 99$ if I wanted to debug my app on the real device. Pay just to debug?! Also I had a doubt about the possibility to test multi-touch gestures with the simulator but it seems it’s possible.
What we can do is the following: you can test it on iOS and I will test on Android. Most of the code will be common to both platforms anyway.
Will create a new repository on bitbucket and give you write persmission asap
I tried a few things already just to check if things are possible:
– Made simple html app with cordova, angularjs and hammer.js and deployed to android phone => it works, should also work on ios devices
– Made a free account on Openshift and made a simple python app with bottle and serve some image urls in json format => works also and can fetch urls from android phone and display images this way.
It seems cron jobs are also possible on openshift so looks good and didn’t find any limitation of read/writes, just limited storage (1GB), memory (512MB) and number of apps: https://www.openshift.com/products/pricing
Also gave you access to the project here: https://bitbucket.org/zeli47/kpop-image-reader
P.S: server code is on a git repo on openshift servers, can give you access to it but need your ssh public key
I use Arch Linux so that’s the one I’m most familiar with but Ubuntu is fine too. Though, not sure we need to do this, you can just install python and “pip” and then install bottle and other dependencies (just run “pip install bottle”), and run embedded dev server for testing on your local machine and then deploy on openshift. Openshift offers easy deployment, no need to mess with server config, just push commits and it auto deploys the app. Downside is collaboration features are not free but hopefully we don’t need them
I just created a python app, and added mysql and phpmyadmin. there are other solutions available if needed, just have a look on their site
It’s optional to install it, you can do almost everything from the website (create app, copy/paste ssh public key there, etc), then use git to push code, etc.
But if you install the rhc gem then it can help you setup your account and upload the ssh key and it checks you have everything installed correctly so maybe it’s a good idea to install it. When it’s installed, just run “rhc setup” and follow the instructions
First you should choose what database to use. On openshift you have 3 types ready for use:
– MySQL
– MongoDB
– PostgreSQL
Pick one and install the “cartridge” for it on openshift website.
Then you have to install a python module to use it, for instance for MongoDB you should install the pymongo module.
Then you can just “import pymongo” and start using it.
Of course you can use mysql or postgresql if you prefer.
Nice tutorial for mongodb with python: http://api.mongodb.org/python/current/tutorial.html
Not sure one or the other will be faster, I think they are both good. The main difference is how the data is stored and how you access it.
With mysql you have tables and rows and it can become quite complicated if you have lots of tables you have to join them and make very long and complicated queries.
With mongodb you have collections of json objects and the structure of the objects can change (with mysql it’s fixed once you create the table).
If you know MySQL already, maybe start with that, or if you want to try something new and don’t mind to spend a little extra time learning, you can try mongodb. You can always change later if it’s not what you wanted 😉
Maybe take 5 minutes to try it out: http://try.mongodb.org/
and see if it’s what you need or not.
KAHI KAHI KAHI 😀
SNSD(Choi Soo-young)
I don’t have an iPad 🙁
Yoon Bo Mi(A Pink).
Это же Боми!? правильно по-верхнему скрину, вроде как…
1.- Mamamoo Solar
2.- Sonamoo Na Hyun
3.- Hello Venus Nara
I’ve seen that pictorial several times everywhere before to not know who they are
Amy winehouse
Muchas gracias
easy one…guess who? красивое фото.

There is an ipad app?
There was. It is out of the store for a few yrs, i still use it as an rss reader (for new pics) , but seems no one else 😎
If you need it , i think i could send it somehow (with dev test signing)
So its RSS reader. I don’t think I need it. When you say “for new pics” what do you mean? twitter?
Originally it is a guess who game with random artists pics (from the set of names)
But i updated it to has another mode of continual pics showing from the selected source (full screen no text) like idol-grapher and Fiddle.se
I plan to introduce the web version (like ex google-reader) but already a few yrs can’t finish it (time always go to somth other)
From comments:
Sources
> Where do we get the images from?
currently i use just google-search (to get urls)
RSS sure will be addded
then it could be useful to add the whole site
(like idol-grapher/fiddle/etc) which have a lot of pics of our interest which we can range by idol/band and time-stamp
> Sites without RSS feeds?
yes this is the problem, how to manage em
> Twitter?
don’t think, maybe there are good chanels (i don’t know such ones, do you?)
> Default sources?
i mean we can pre-fill some sources (where to take good pics from):
– fiddle
– i-grapher
– 4sashi.com 8)
> Custom sources?
yes, sources could be editable
to add new ones, remove default ones,
+in settings we can
– add/remove idols/bands
– set modes: random on-off, mid quality, HD, face only, etc.
> How do custom sources deal with historical data?
same 8)
storing all from adding time point for RSS
and parse site (or use google-search) for none-RSS
Storage
> Image URL?
yes
> Image metadata (date, tags)
don’t know
> Image itself?
no