I'm cataloging feedback I receive so you all know I'm listening, and will be iteratively adding these items as we move forward over the next week or two.
You can find the list on the post below. I'll be updating this post in real time from feedback I get here, on Twitter, or through Intercom (blue button on site).
Honestly I'm not really sure how this works - but uhh.. maybe the Fork 'n Go concept could be applied for more "out of the box" ease? (https://github.com/jlord/forkngo)
I can't help but think at some point GitHub will not be so friendly to free blog hosting through their system. They are a version control SAAS. If you generate to much traffic, what will happen? The market might be ripe for a git based cms/blog why not build that company and service?
Really interesting point. I'm not exactly sure what I think would happen, but my intuition is that GitHub wouldn't turn off access to GitHub pages just for users on our platform - even if traffic was excessive. I think this generates a lot of interest in GitHub in general which is beneficial for them.
But you are right, I believe a git based cms/blog that runs in some standalone fashion could be a v2 or v3 of this product. Or perhaps a one-off paid option if you don't want to use GitHub, you can pay to use our private git repo or even a git repo you host on your own.
Solution: Host the blog on your own domain name. That's literally all you need to ensure that down the line you can move the content to your own jekyll infrastructure (or somebody else's).
It's true. At some point github will turn all this blog traffic into profit. But just like they did with gist.github.com they can make a ghpages.github.com or something and create plans. At this point I think is just another feature that keep them a big step ahead of bitbucket, gitlab, and others.
> The market might be ripe for a git based cms/blog why not build that company and service?
But what's a `git based cms/blog' ? A static website generator that commit on heroku ? Or a cli/gui post manager that uses git command-line argument as an API ?
I agree completely. I'm working on a static website CMS that I hope to release soon. I think there's a lot of room for the static website market to grow, particularly alongside proper management and deployment tools.
Really good suggestions. I appreciate the feedback. I am definitely going to increase the quality of the gif, and figure out a way to include a demo of the product.
You're absolutely right. Even me being the creator of this the one vibe I get when using the product is "well this is not very intuitive". I'm going to be optimizing for that over the next few weeks. Thanks for the gut check.
I'm not sure if it's possible or not but you may try to create an organization for your application and create repositories in that organization. Maybe you can avoid requesting private repo access using this technique. (Mixpanel uses a similar technique https://github.com/orgs/mixpanel-platform)
A feature where it creates pull-requests would even remove the need for it to have any access. Not sure if it's a good idea, as it adds more steps (and no post-in-future). Or maybe just add it as a collaborater to the project could work. I don't have a problem with how it works now, though.
Is it possible to use TinyPress with already modified Jekyll?
I already host a blog on github and am using jekyll. What I would like to do is take advantage of that neat backend for writing posts.
Interesting. I'm re-building the core engine powering this using (basically all Javascript right now on Backbone) in Ruby + Javascript with an actual postgresql database behind it. Right now it's using an in memory store for things like your autosave drafts etc.
Do you think I should open source the project? Right now the Android app and the web-app (mentioned above) are closed source, but the philosophy of this platform is about open source writing - and to not have an open source application powering that seems antithetical. What do you think?
I would say it depends on your personal goals. If you're doing this as you say to open source wiring and not the application then there's your answer. If you'd like to see it develop more than you have time for theb maybe open source would be a better option or talk to some friends that find it interesting and build it together white keeping it closed. So weigh your options and what you'd like to see happen to it and go from there.
Open source would be great! Sounds like you've already got some prospective contributors out there, and an open source project will naturally be easier to market to GitHub users.
Yep. A verry rough v1 of this has been in the works for quite some time, with a few friends of mine. But they have all moved on to other projects and this is now my baby. Also, this is a completely overhauled version - hence today's HN post.
I'm cataloging feedback I receive so you all know I'm listening, and will be iteratively adding these items as we move forward over the next week or two.
You can find the list on the post below. I'll be updating this post in real time from feedback I get here, on Twitter, or through Intercom (blue button on site).
https://tinypressco.github.io/2015/10/18/on-hacker-news.html