5 ms·

Ask HN: How long would it take you to build instagram?

14 points by amerf1 14y ago

How long would you take? how much help? what platform?

c1sc0 14y ago
Check out AnyPic: https://parse.com/tutorials/anypic https://parse.com/tutorials/anypic
- gghootch 14y ago
  With the help of Parse's platform and this anypic tutorial/template I went from 'I built a simple Ruby web service once' to having created a fully customized Instagram-like service in about three months of 30% of my time spent on development. In the end, virtually everything was replaced and/or altered.
  - moconnor 14y ago
    And then what happened to it? What did you do with it?
    - gghootch 14y ago
      It was for a frighteningly ambitious side-project initiated thanks to a great university course - http://frapp.it http://frapp.it Unfortunately, due to time constraints we went the built first, talk to mentors etcetera and only then do actual diligent market research route. Everyone we've spoken to recognizes some value in the project and came up with different USPs. As a result, focus shifted dramatically in a matter of weeks, ultimately leading to a complete lack of focus and loss of a common goal. This week we re-evaluated it all, decided to work out these wrinkles in the coming weeks and start selling aggressively to a narrow collection of potential customers. When all is said and done, I expect the iOS app will be shelved. Did learn a lot about iOS development, though :D!
iurisilvio 14y ago
What exactly is Instagram? I can create a cool way to upload photos, add some filters and things like that. It is easy. Instagram value is not only about software. You have a big community, integration with a lot of services, a lot of servers to handle the stupid amount of images uploaded in a day. So, it is not probably about days. If you want to create the next Instagram, you should do something far better than that.
graue 14y ago
So I don't know how long it will take but one of my side project ideas is to build an Instagram-like photo sharing social network that uses your Dropbox for storage. Since a lot of my ideas turn out to be unoriginal, does anyone know if this already exists? All I can find so far are apps and IFTTT recipes to back up your Instagram photos to Dropbox, e.g. https://instadrop.appspot.com/ https://instadrop.appspot.com/. But what appeals to me here is the idea of decoupling the app (photo sharing with filters) from the storage (Dropbox). I think that might give users more control and create new possibilities.
- jmathai 14y ago
  Not really an Instagram clone but OpenPhoto has filters in the iPhone app and let's you store photos in Dropbox. http://theopenphotoproject.org http://theopenphotoproject.org https://github.com/photo https://github.com/photo (for all the goodies) Disclaimer: I'm the lead dev.
  - graue 14y ago
    Does it have a social aspect? The pitch seems to be all "you ... your photos ... you decide ... your photos", with no mention of sharing. What I really have in mind is a way to share photos (follow, be followed, comment), not just upload them for safekeeping, which I'd guess is probably a use-case far fewer people are interested in. Again, I'm just responding to the home page and "overview" subpage... if there is a social aspect your team should mention it more prominently.
    - jmathai 14y ago
      We haven't been very focused on sharing. We do let you post out to Twitter and Facebook though. But it's not replacement for the Flickr or Instagram communities. We've been a little slower on adding social features right in the platform. With very limited resources it just isn't high on our priority list. On a side note, I'm not convinced the size of the sharing market is as big and can be sustained as many people think (besides Facebook). I think the vast majority of photos which people place a lot of value on are ones they share with specific family and friends (this excludes photos of sunsets, lattes, and cats). The safekeeping of these photos is a big deal. There's a lot going on with private social networks which I think is interesting but the chicken/egg problem for small networks is BIG. ANYWAYS, the project is open sourced for a reason. We've built and are focused on the platform and safekeeping of photos. Pure social features may or may not find their way in but anyone can fork and add it. We provide a very stable and robust base to work off of.
BarkMore 14y ago
There are articles on the web about Instagram's history. If I recall correctly, a couple of people built the first version in a small number of months including both server and client. The server used Python, PostgreSQL, Redis and Linux.
- webology 14y ago
  FWIW the server is / was written in Django.
Vitaly 14y ago
From our experience whatever the service you are building it should usually take somewhere in the area of 200-500 hours for the first public version. If/when it catches up it will take thousands of hours to get it to a polished version if extras.
herval 14y ago
Less than a month, possibly (or a bit more, if you plan on prematurely-optimize it for instagram-level traffic - which will possibly never happen for the "clone" anyway). The hard part, really, would be BECOMING Instagram. The software is pretty much trivial (as was Facebook at the beginning)...
alexhjones 14y ago
How much testing would you want to do as part of 'build'?
hboon 14y ago
3-12 man months, assuming just 1 mobile front end. Like most software, it helps so much if you know exactly what you want to build and can plan ahead and not need to make too many changes before you even ship. Constraining yourself to the smallest reasonable feature set helps you ship. Don't forget, you need to build both the frontend and backend. Software like this needs to be as asynchronous as possible and ready to fan-out, even if you know you wouldn't need to scale to 10M users from day 1. Use as many pre-built and vanilla UI components as possible. Custom UI slows down development by a lot.
onlyup 14y ago
A year or two.

14 points by amerf1 14y ago

How long would you take? how much help? what platform?

c1sc0 14y ago

Check out AnyPic: https://parse.com/tutorials/anypic https://parse.com/tutorials/anypic

gghootch 14y ago
With the help of Parse's platform and this anypic tutorial/template I went from 'I built a simple Ruby web service once' to having created a fully customized Instagram-like service in about three months of 30% of my time spent on development. In the end, virtually everything was replaced and/or altered.
- moconnor 14y ago
  And then what happened to it? What did you do with it?
  - gghootch 14y ago
    It was for a frighteningly ambitious side-project initiated thanks to a great university course - http://frapp.it http://frapp.it Unfortunately, due to time constraints we went the built first, talk to mentors etcetera and only then do actual diligent market research route. Everyone we've spoken to recognizes some value in the project and came up with different USPs. As a result, focus shifted dramatically in a matter of weeks, ultimately leading to a complete lack of focus and loss of a common goal. This week we re-evaluated it all, decided to work out these wrinkles in the coming weeks and start selling aggressively to a narrow collection of potential customers. When all is said and done, I expect the iOS app will be shelved. Did learn a lot about iOS development, though :D!

iurisilvio 14y ago

What exactly is Instagram? I can create a cool way to upload photos, add some filters and things like that. It is easy. Instagram value is not only about software. You have a big community, integration with a lot of services, a lot of servers to handle the stupid amount of images uploaded in a day. So, it is not probably about days. If you want to create the next Instagram, you should do something far better than that.

graue 14y ago

So I don't know how long it will take but one of my side project ideas is to build an Instagram-like photo sharing social network that uses your Dropbox for storage. Since a lot of my ideas turn out to be unoriginal, does anyone know if this already exists? All I can find so far are apps and IFTTT recipes to back up your Instagram photos to Dropbox, e.g. https://instadrop.appspot.com/ https://instadrop.appspot.com/. But what appeals to me here is the idea of decoupling the app (photo sharing with filters) from the storage (Dropbox). I think that might give users more control and create new possibilities.

jmathai 14y ago
Not really an Instagram clone but OpenPhoto has filters in the iPhone app and let's you store photos in Dropbox. http://theopenphotoproject.org http://theopenphotoproject.org https://github.com/photo https://github.com/photo (for all the goodies) Disclaimer: I'm the lead dev.
- graue 14y ago
  Does it have a social aspect? The pitch seems to be all "you ... your photos ... you decide ... your photos", with no mention of sharing. What I really have in mind is a way to share photos (follow, be followed, comment), not just upload them for safekeeping, which I'd guess is probably a use-case far fewer people are interested in. Again, I'm just responding to the home page and "overview" subpage... if there is a social aspect your team should mention it more prominently.
  - jmathai 14y ago
    We haven't been very focused on sharing. We do let you post out to Twitter and Facebook though. But it's not replacement for the Flickr or Instagram communities. We've been a little slower on adding social features right in the platform. With very limited resources it just isn't high on our priority list. On a side note, I'm not convinced the size of the sharing market is as big and can be sustained as many people think (besides Facebook). I think the vast majority of photos which people place a lot of value on are ones they share with specific family and friends (this excludes photos of sunsets, lattes, and cats). The safekeeping of these photos is a big deal. There's a lot going on with private social networks which I think is interesting but the chicken/egg problem for small networks is BIG. ANYWAYS, the project is open sourced for a reason. We've built and are focused on the platform and safekeeping of photos. Pure social features may or may not find their way in but anyone can fork and add it. We provide a very stable and robust base to work off of.

BarkMore 14y ago

There are articles on the web about Instagram's history. If I recall correctly, a couple of people built the first version in a small number of months including both server and client. The server used Python, PostgreSQL, Redis and Linux.

webology 14y ago
FWIW the server is / was written in Django.

Vitaly 14y ago

From our experience whatever the service you are building it should usually take somewhere in the area of 200-500 hours for the first public version. If/when it catches up it will take thousands of hours to get it to a polished version if extras.

herval 14y ago

Less than a month, possibly (or a bit more, if you plan on prematurely-optimize it for instagram-level traffic - which will possibly never happen for the "clone" anyway). The hard part, really, would be BECOMING Instagram. The software is pretty much trivial (as was Facebook at the beginning)...

alexhjones 14y ago

How much testing would you want to do as part of 'build'?

hboon 14y ago

3-12 man months, assuming just 1 mobile front end. Like most software, it helps so much if you know exactly what you want to build and can plan ahead and not need to make too many changes before you even ship. Constraining yourself to the smallest reasonable feature set helps you ship. Don't forget, you need to build both the frontend and backend. Software like this needs to be as asynchronous as possible and ready to fan-out, even if you know you wouldn't need to scale to 10M users from day 1. Use as many pre-built and vanilla UI components as possible. Custom UI slows down development by a lot.

onlyup 14y ago

A year or two.