![Introducing Artifact](docs/data/artifact-thumb.png)
- Installation Guide
- Quick Start Guide
- Cheat Sheet
- FAQ
- Simple Quality Book
- Rendered Design Documents
- Issue Tracker
Artifact is the simple, linkable and trackable design documentation tool for everybody. It allows anyone to write and link their design documents both to each other and to source code, making it easy to know how complete their project is. Documents are revision controllable, can be edited in the browser and have a full suite of command line tools for searching, displaying, checking, exporting and formatting them.
![b0rk scenes from writing design docs](docs/data/attribution/b0rk-design-documents.jpg-large)
Writing detailed design documents is one of the core pillars of quality software development. Design documents are how you capture the requirements (purpose) of your project and link them to your specifications (how you will build it). They let you get your ideas on paper before writing code, and help you have fewer painful refactors. They create a reference for developers and curious users of how and why your project was developed a certain way, and make it easier to refactor your project when that becomes necessary.
Even though design documents are critical to the quality of software, there are very few tools for writing them and integrating them into the larger context of a project. Artifact aims to fill the major gap in quality best practices by making writing good design documents useful to the average developer.
First of all, artifact makes it easy to write design documents in text files
and link them by just specifying their partof
attribute. This allows
developers to put their design documents under revision control, review them
using regular code review tools and use all the normal text processing tools
(vim, grep, sed, etc) to view, edit and refactor them. Artifact also provides
some command line tools of its own.
Secondly, design documents can be linked to source-code through a language
agnostic syntax (#ART-name
anywhere in the source code). Once linked, anyone
reading the comment can easily look up the relevant design documents. In the
same way, anyone looking at the Web UI can see exactly where a specification or
test is implemented in code. Furthermore, if the name of a design doc changes,
art check
will tell you where your dangling references are. Never again will
you have to be scared of refactoring your design documents because your
references in code will be out of date.
Finally, artifact exports a beautiful rendered view of your design documents
onto sites like github-sites (example) and you can edit in your browser
using art serve
. This completes the self documenting nature and allows
anyone (even non-developers!) to view and edit the design documents of their
project.
In this way, artifact aims to unify other quality best practices while also make writing design documents more fun and useful in your day to day development efforts.
Jump into artifact with the the youtube commercial and the Quick Start Guide.
You can support the project by โญ staring it on github and ๐ sharing it with your friends, coworkers and social media. You can also support it directly on patreon, vote for it in the 2017 Hackaday Prize and by leaving direct feedback.
All funds collected through any link above will only go towards hiring student developers as interns to work on open source features within artifact. No money from these sources will go to myself (Garrett Berg) or to the development of proprietary software.
By supporting artifact, you are supporting open source tools for quality software development and also internship oportunities for students passionate about open source. Thank you!
Artifact is 1.0 software with a strong commitment to backwards compatibility. The 1.0 release is the "open source" release. Artifact is ready for projects of any size to use it as their defacto design documentation tool, but the external tooling may still be lacking for enterprise customers.
The 2.0 release will focus on stabilizing the library for external tooling. This will position artifact for integration with industry tools such as JIRA and external regression test tracking software. I am currently seeking enterprise support, please consider supporting this project on patreon.
The following are stable APIs that should always remain backwards compatible:
- Artifact
.toml
files. Features may be added, but should not be removed. This includes:- Artifact types:
REQ
,SPC
,TST
. - Artifact fields:
partof
,text
,done
. - Text format: markdown by default with
[[ART-name]]
links.
- Artifact types:
- Artifact relationships: explicit partof, auto linking by name, etc.
- Artifact completeness calculated from its relationships.
- Source code links:
#ART-name
anywhere in the source code marks the artifact as done. .art/settings.toml
file- The command line interface, including:
- The name of the commands (
tutorial
,init
,ls
,check
,fmt
, etc). - Existing flags for each command.
- The functional checks that are completed (i.e. dangling artifact names)
art serve
default port of 5373 on localhost
- The name of the commands (
The following should remain relatively stable but may have minor tweaks before 2.0:
- The output of commands, including:
- The format of the output messages. I.e. you should not rely on
art ls
having a specific output format. - The format of artifacts as performed by
art fmt
.
- The format of the output messages. I.e. you should not rely on
- The json format returned by
art ls --json
and through the json-rpc server fromart serve
. - Anything not mentioned in the first section. If you are unsure, please open a ticket.
The following are expected to change a lot before 2.0:
- The web ui. Hopefully the changes will be an almost uniform improvement for everybody.
- The
art serve
http interface including its API methods and data format. - The code and expected functions/types/etc of the library itself should be considered highly unstable. Future work will involve breaking it into smaller crates that are more stable.
- Logging messages evoked with
art -v
. - Lots of other things that I can't think of. If you are unsure or concerned, open a ticket.
Artifact will continue to use a continuous-release cycle with extensive unit and integration tests. There will also be a beta release channel for new and experimental features. If you find a bug please open a ticket.
The intent of the artifact licensing is that:
- The artifact application be open source for anyone to use for any purpose (dual licensed MIT + Apache).
- Anything generated by artifact can be licensed any way the user wishes to, including the built static html pages.
- Any documents, tutorials or specifications for artifact (except the code and logo) remain public domain and can be used for any purpose at all.
All documentation and tutorials for the artifact application are released under the CC0 Creative Commons Public Domain License with the intent that you should feel free to copy, paste and modify any of the designs, guides examples or exported data for any purpose (including commercial) without the need of attribution. You can read more about CC0 here: https://creativecommons.org/publicdomain/
The CC0 license applies to:
- All project docs
- The Artifact Design Documents
(also located in
design/
) - The Artifact Documentation (located in
docs/
) except the logo indocs/logo
- The Artifact Wiki
- Any file or data created by any artifact command, including:
- documents created by
art init
- documents created by
art tutorial
- compiled html/css/json files created by
art export
- documents created by
The artifact logo (named Tula) is licensed under Creative Commons
Attribution-ShareAlike (CC BY-SA
) and can be used by the artifact project for
any purpose without needing additional attribution. The artifact logo is located
in docs/logo
and was originally created by
packapotatoes.
The source code is Licensed under either of
- Apache License, Version 2.0, (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
- MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)
at your option.
Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.