Fossil

Questions And Criticisms
Login

Questions And Criticisms

This page is a collection of real questions and criticisms that were raised against Fossil early in its history (circa 2008). This page is old and has not been kept up-to-date. See the change history of this page.

Fossil sounds like a lot of reinvention of the wheel. Why create your own DVCS when you could have reused mercurial?

I wrote fossil because none of the other available DVCSes met my needs. If the other DVCSes do meet your needs, then you might not need fossil. But they don't meet mine, and so fossil is necessary for me.

Features provided by fossil that one does not get with other DVCSes include:

  1. Integrated wiki.
  2. Integrated bug tracking
  3. Immutable artifacts
  4. Self-contained, stand-alone executable that can be run in a chroot jail
  5. Simple, well-defined, enduring file format
  6. Integrated web interface
Why should I use this rather than Trac?
  1. Fossil is distributed. You can view and/or edit tickets, wiki, and code while off network, then sync your changes later. With Trac, you can only view and edit tickets and wiki while you are connected to the server.
  2. Fossil is lightweight and fully self-contained. It is very easy to setup on a low-resource machine. Fossil does not require an administrator.
  3. Fossil integrates code versioning into the same repository with wiki and tickets. There is nothing extra to add or install. Fossil is an all-in-one turnkey solution.
Love the concept here. Anyone using this for real work yet?
Fossil is self-hosting. In fact, this page was probably delivered to your web-browser via a working fossil instance. The same virtual machine that hosts http://fossil-scm.org/ (a Linode 720) also hosts 24 other fossil repositories for various small projects. The documentation files for SQLite are hosted in a fossil repository here, for example. Other projects are also adopting fossil. But fossil does not yet have the massive user base of git or mercurial.
Fossil looks like the bug tracker that would be in your Linksys Router's administration screen.

I take a pragmatic approach to software: form follows function. To me, it is more important to have a reliable, fast, efficient, enduring, and simple DVCS than one that looks pretty.

On the other hand, if you have patches that improve the appearance of Fossil without seriously compromising its reliability, performance, and/or maintainability, I will be happy to accept them. Fossil is self-hosting. Send email to request a password that will let you push to the main fossil repository.

It would be useful to have a separate application that keeps the bug-tracking database in a versioned file. That file can then be pushed and pulled along with the rest repository.

Fossil already does push and pull bugs along with the files in your repository. But fossil does not track bugs as files in the source tree. That approach to bug tracking was rejected for three reasons:

  1. Check-ins in fossil are immutable. So if tickets were part of the check-in, then there would be no way to add new tickets to a check-in as new bugs are discovered.
  2. Any project of reasonable size and complexity will generate thousands and thousands of tickets, and we do not want all those ticket files cluttering the source tree.
  3. We want tickets to be managed from the web interface and to have a permission system that is distinct from check-in permissions. In other words, we do not want to restrict the creation and editing of tickets to developers with check-in privileges and an installed copy of the fossil executable. Casual passers-by on the internet should be permitted to create tickets.

These points are reiterated in the opening paragraphs of the Bug-Tracking In Fossil document.

Fossil is already the name of a plan9 versioned append-only filesystem.
I did not know that. Perhaps they selected the name for the same reason that I did: because a repository with immutable artifacts preserves an excellent fossil record of a long-running project.
The idea of storing a repository in a binary blob like an SQLite database terrifies me.
The use of SQLite to store the database is likely more stable and secure than any other approach, due to the fact that SQLite is transactional. Fossil also implements several internal self-checks to insure that no information is ever lost.
I am dubious of the benefits of including wikis and bug trackers directly in the VCS - either they are under-featured compared to full software like Trac, or the VCS is massively bloated compared to Subversion or Bazaar.

I have no doubt that Trac has many features that fossil lacks. But that is not the point. Fossil has several key features that Trac lacks and that I need: most notably the fact that fossil supports disconnected operation.

As for bloat: Fossil is a single self-contained executable. You do not need any other packages (diff, patch, merge, cvs, svn, rcs, git, python, perl, tcl, apache, sqlite, and so forth) in order to run fossil. Fossil runs just fine in a chroot jail all by itself. And the self-contained fossil executable is much less than 1MB in size. (Update 2015-01-12: Fossil has grown in the years since the previous sentence was written but is still much less than 2MB according to "size" when compiled using -Os on x64 Linux.) Fossil is the very opposite of bloat.