4 ms·

This is why I find it really difficult to adopt smaller products. They quite frequently get shut down.

DoofusOfDeath 8y ago
I agree, and it creates a real chicken-and-egg problem. I've also been on the other side of things, working for such companies. I've always felt a bit of unease knowing that prospective customers may be taking on more risk than our salespeople were letting on.
- JMTQp8lwXL 8y ago
  I guess there's people willing to take that risk, or have short-term benefit of whatever your snowflake solution of a service is, that nobody else could provide for you-- e.g., use a service provided by small business for 2 years or have no solution whatsoever. I don't know Waffle very well, but if it's just some project management thing, chances are there's a more mature app that does 90%+ of that.
bachmeier 8y ago
It would be a lot easier to invest in a small company if there was a guarantee that the project would be made open source if the company doesn't make it. Unfortunately, that rarely happens, and then you're stuck. (Doesn't even have to be a small company - people have been worried Evernote for years.)
ken 8y ago
The primary issue is the cost of switching, and Waffle was actually very good at this. It was basically just a front-end for GitHub Issues. I always knew that if at any point Waffle shut down, I'd be no worse off than before I started using them, and I'd still have all my data. I'm actually much more concerned about GitHub itself. I still haven't found a good way to back up my Issues. They used to point to joeyh's Haskell program, which is neat but has many severe limitations (like: doesn't support private repos at all). Their current recommendation seems to be a third-party service called "BackHub", whose cheapest plan is more expensive than GitHub Pro.
- JMTQp8lwXL 8y ago
  Maybe issues should just be built into Git itself. Can't really imagine software development without them.
  - xkcd-sucks 8y ago
    If you track issues in a text file, sqlite DB etc -- they already are in git. (But I've never seem anyone do this)
    - ken 8y ago
      SQLite files can't easily be merged. Putting one database in another database doesn't allow you to take advantage of the best features of either one. Text files are a good idea, and I've done that for many of my own projects. You just need to figure out a filename scheme, and a file format (JSON? YAML?). It's great to be able to atomically commit a fix to the source code, and also close the bug report. I'm not sure how well it would scale. (Mozilla has 1.5M bug reports in their Bugzilla!) At the least, you'd need a local client that could index the text files into a format more suitable for sorting/searching. It would only need to download the diffs, though.
      - xkcd-sucks 8y ago
        Oh oops, there already exist bug trackers for a few decades :) Take your favorite bug tracker and put its offline storage into a git submodule?? I imagine they are all better than GitHub issues which are literally just a forum of threads

I agree, and it creates a real chicken-and-egg problem. I've also been on the other side of things, working for such companies. I've always felt a bit of unease knowing that prospective customers may be taking on more risk than our salespeople were letting on.

JMTQp8lwXL 8y ago
I guess there's people willing to take that risk, or have short-term benefit of whatever your snowflake solution of a service is, that nobody else could provide for you-- e.g., use a service provided by small business for 2 years or have no solution whatsoever. I don't know Waffle very well, but if it's just some project management thing, chances are there's a more mature app that does 90%+ of that.

bachmeier 8y ago

It would be a lot easier to invest in a small company if there was a guarantee that the project would be made open source if the company doesn't make it. Unfortunately, that rarely happens, and then you're stuck. (Doesn't even have to be a small company - people have been worried Evernote for years.)

ken 8y ago

The primary issue is the cost of switching, and Waffle was actually very good at this. It was basically just a front-end for GitHub Issues. I always knew that if at any point Waffle shut down, I'd be no worse off than before I started using them, and I'd still have all my data. I'm actually much more concerned about GitHub itself. I still haven't found a good way to back up my Issues. They used to point to joeyh's Haskell program, which is neat but has many severe limitations (like: doesn't support private repos at all). Their current recommendation seems to be a third-party service called "BackHub", whose cheapest plan is more expensive than GitHub Pro.

JMTQp8lwXL 8y ago
Maybe issues should just be built into Git itself. Can't really imagine software development without them.
- xkcd-sucks 8y ago
  If you track issues in a text file, sqlite DB etc -- they already are in git. (But I've never seem anyone do this)
  - ken 8y ago
    SQLite files can't easily be merged. Putting one database in another database doesn't allow you to take advantage of the best features of either one. Text files are a good idea, and I've done that for many of my own projects. You just need to figure out a filename scheme, and a file format (JSON? YAML?). It's great to be able to atomically commit a fix to the source code, and also close the bug report. I'm not sure how well it would scale. (Mozilla has 1.5M bug reports in their Bugzilla!) At the least, you'd need a local client that could index the text files into a format more suitable for sorting/searching. It would only need to download the diffs, though.
    - xkcd-sucks 8y ago
      Oh oops, there already exist bug trackers for a few decades :) Take your favorite bug tracker and put its offline storage into a git submodule?? I imagine they are all better than GitHub issues which are literally just a forum of threads