4 ms·

Normal yes, not much beneficial. I dont have this problem at all (take a look at my long list of Show HN of all kinds, including several very poor half-baked th

by soneca 10y ago

Normal yes, not much beneficial. I dont have this problem at all (take a look at my long list of Show HN of all kinds, including several very poor half-baked things that I'm not that proud of), so I dont think I can give any empathically useful advice. But I would love to know, what are you building? (Who knows, maybe it does indeed requires a longer gestation period).

fratlas 10y ago
ML-based social platform where the algorithms learn a user's tastes. Limited to images, it's somewhere between Tumblr/VSCO/Pinterest/IG. It works for me, and my girlfriend loves using it, but the problem is she always wants to export her chosen images back to another platform for posting. I sense it will be a chicken and the egg problem. Was mostly so I could learn how to handle big data (~1B edges)
- soneca 10y ago
  Well, the discovery of new images to share on your preferred social media is a problem that some people have. Why dont you embrace it instead of fighting it and trying to build your own social media?
  - fratlas 10y ago
    Indeed, I built it for my own needs in the first place. I've collected the data from multiple sources, some of the TOS I agreed to explicitly said not to use it in competition as a social media platform. I would be happy enough if people used it for discovery.
    - soneca 10y ago
      so ship it to the wild and see how they use it! :)
- NumberCruncher 10y ago
  No product survives the first contact with the customer. This already happened to you despite of having only one customer - your girlfriend. If I were you I would be really happy about her wanting to post the chosen images back to an other platform. This means your product would go viral on its own without any help. Think of a fair annual price, triple it and go live. There are many bored people out there with to much money in their pockets.
  - fratlas 10y ago
    I suppose you are right. At the very least it's an item for the resume.

by soneca 10y ago

fratlas 10y ago

ML-based social platform where the algorithms learn a user's tastes. Limited to images, it's somewhere between Tumblr/VSCO/Pinterest/IG. It works for me, and my girlfriend loves using it, but the problem is she always wants to export her chosen images back to another platform for posting. I sense it will be a chicken and the egg problem. Was mostly so I could learn how to handle big data (~1B edges)

soneca 10y ago
Well, the discovery of new images to share on your preferred social media is a problem that some people have. Why dont you embrace it instead of fighting it and trying to build your own social media?
- fratlas 10y ago
  Indeed, I built it for my own needs in the first place. I've collected the data from multiple sources, some of the TOS I agreed to explicitly said not to use it in competition as a social media platform. I would be happy enough if people used it for discovery.
  - soneca 10y ago
    so ship it to the wild and see how they use it! :)
NumberCruncher 10y ago
No product survives the first contact with the customer. This already happened to you despite of having only one customer - your girlfriend. If I were you I would be really happy about her wanting to post the chosen images back to an other platform. This means your product would go viral on its own without any help. Think of a fair annual price, triple it and go live. There are many bored people out there with to much money in their pockets.
- fratlas 10y ago
  I suppose you are right. At the very least it's an item for the resume.