Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

From the research I have done there are three major types of git server buyers:

1) We are so small that we don't even think twice and buy some private repos on GitHub. (i.e. no market there)

2) We are so big that we can't possibly even fathom the idea of having our code on someone else's server so GitHub.com is out of the question. Sadly GitHub:Fi is per/user and because we are really big that means we have lots of devs so it is too expensive in most cases. (i.e. enterprise sale yuck)

3) For medium sized companies that either have some devs (more then 5) or a ton of repos or large repos they want an alternative to github. Many are slapping up gitolite which while good is a far cry from github.

Like I mentioned I already have found companies that want this solution and they are using GitHaven. The real question is if this market is too small or should I just move on to another project, open source GitHaven and let everyone (who wants to) slap it up as a frontend to their existing gitolite installs?



Most products targeted at developers are per-seat priced (or have pricing derived from per-seat pricing); it's how you do value pricing when value scales with the number of developers. Look at how Atlassian does it; that's still effectively per-seat licensing (it requires prospects to think about how many devs they'll have).

You can try to avoid these schemes, but you have to do it and not go out of business.


Yeah the per-seat point is really a moot point when it comes to enterprise. ClearCase is $4000 a license and a lot and I mean a lot of large companies use it. I know of a company that is currently spending $2.5 million a year on ClearCase licenses and they really want to get off it but ClearCase is so entrenched/integrated into their work flow that this is not a trivial task. Github is not going to solve this problem for them.

The problem with Github:Fi is the value proposition is extremely weak when it comes to meeting enterprise needs.

- They have an issue tracker that doesn't support custom fields. This is critical for companies with 500+ developers and/or very diverse product lines.

- They have a wiki but you can get confluence for $12,000 for the entire company which is more feature rich. It has decent searching.

- They have blogging but this is of no value in the enterprise world. The majority of employees really don't have the time or care about blogging internally.

- They have Gist which offers no real value for the same reason why blogging isn't that big of a deal.

When it's all over, their only strong value proposition for the enterprise world is their intuitive push and pull interface. But the problem with this is they are competing directly with free solutions like:

- gitorious

- gitolite + internal web interface

- gerrit

I'm obviously speculating here, but based on how their pricing model changed:

old: 1000 per user (initial purchase) + 200 something a year per user subscription cost

new: 250 per user if less than 25 users and 200 per user if more than 25 users

I think they realized they over estimated their value proposition initially. If Github wants to seriously break into the enterprise world, they are either going to have to lower their per seat cost and/or strengthen their value proposition.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: