1-pager draft proposal
Summary
Adds a general purpose, high level proposal intended to establish the broad need for a campus-level git service. Takes benefit to programmers as a given, and focuses instead on value to students and faculty. Does not cover implementation details, which would be handled in a future phase/proposal. Intended audience is department-level decision makers who could authorize further investigation.
Review needs
- Impact
- Does the proposal make a convincing case that a shared git service would be beneficial to students, faculty, and staff?
- Can assertions be backed up by faculty statements or public data?
- Wordsmithing
- Check for readability
- Minimize jargon
- Add a snappy title
- Sources
- I lean heavily on github's blogs. Can sources be diversified?
- Do footnotes match references appropriately?
Merge request reports
Activity
- Proposals/campus-git-1-page-proposal.md 0 → 100644
24 25 **Research Support** 26 27 The university research community needs robust version control and collaboration tools, but is often bound by stringent access control requirements (e.g., HIPAA). We have an opportunity to support research innovation by hosting a storage solution tailored to research needs; combining verifiable data storage practices with easy-to-use, cross-institutional team management and visualization tools[13]. 28 29 **Administrative Support & Campus Collaborations** 30 31 The university could realize immediate administrative support savings by selecting an off-the-shelf appliance from a trusted vendor. Campus units are already licensing and running near-identical services for their own use. By committing to the initial outlay and one-time configuration of a centralized service, we reduce not just the up-front costs of duplication, but the hidden wastes of staff efficiency as well. 32 33 ## Footnotes 34 35 1. Version Control Systems Popularity in 2016: https://rhodecode.com/insights/version-control-systems-2016. 36 2. R. Rutenbar, Department Head, Department of Computer Science, University of Illinois at Urbana-Champaign (personal communication, unverified) 37 3. L. Angrave. Senior Lecturer, Department of Computer Science. University of Illinois at Urbana-Champaign (personal communication, December 2, 2015) 38 4. GitHub Education Blog: https://github.com/blog/category/education 39 5. Research IT @ Illinois Training: https://researchit.illinois.edu/training/ This URL seems to have changed to https://researchit.illinois.edu/list-of-available-resources/#training
I can't find a mention of Git training, though. Has that possibly been removed since you included that link?
- Proposals/campus-git-1-page-proposal.md 0 → 100644
1 # Campus Git Service 1-Pager 2 3 ## Summary 4 5 Today's university environment runs on code. From coursework to administration to research and innovation, a robust version control system (VCS) is the bedrock tool in any software development workflow, and git-based version control is the industry standard[1]. Employers expect new Computer Science graduates to have experience using GitHub[2], a leading cloud-based git service. Our academic peers have incorporated git into their classrooms for years[3] [4]. Campus computing organizations offer git training alongside sessions on campus IT orientation, data management, and popular programming languages[5]. Campus IT Pros use git to manage their codebases, but also as a learning tool, sharing code snippets and suggestions in order to benefit from each other's experience. Is it a problem to mention GitHub specifically, since other git implementations like GitLab could also be viable products? Would phrasing along the lines of "Employers expect new Computer Science graduates to have experience using git platforms from vendors like GitHub" accurately represent R. Rutenbar's statement?
Ordinarily I would say yes. In this case, according to @mussulma, he was referring specifically to GitHub.
added 1 commit
- 79581733 - Minor typographical modifications, mostly not changing meaning. Feel free to dis…
- Proposals/campus-git-1-page-proposal.md 0 → 100644
1 # Campus Git Service 1-Pager 2 3 ## Summary 4 5 Today's university environment runs on code. From coursework to administration to research and innovation, a robust version control system (VCS) is the bedrock tool in any software development workflow, and git-based version control is the industry standard[1]. Employers expect new Computer Science graduates to have experience using GitHub[2], a leading cloud-based git service. Our academic peers have incorporated git into their classrooms for years[3] [4]. Campus computing organizations offer git training alongside sessions on campus IT orientation, data management, and popular programming languages[5]. Campus IT Pros use git to manage their codebases, but also as a learning tool, sharing code snippets and suggestions in order to benefit from each other's experience. 6 7 Despite git's ubiquity, the University of Illinois does not offer a campus- or university-level git-based version control service. The closest available option is AITS Subversion[6], which does not support git, nor does it offer the kinds of features commonly found and expected in popular, modern git services[7] [8] [9]. This void has fragmented the campus version control landscape. Though some units do use AITS subversion, many others find this an unsuitable solution. Some units purchase a license and host their own git service, duplicating costs across units; others use cloud-based vendor services, storing university assets offsite; still others, particularly small units and cross-campus affinity groups, simply do without. The net effect is an enormous cost in money, time, and opportunity, with greatly increased business continuity risks. Re: "...storing university assets offsite..." Should we make clearer the ramifications of storing data off-site without proper governance/service agreements/etc? Storing off-site isn't in and of itself the problem, since we may be advocating a cloud-based solution. The real problem is potentially storing sensitive data off-site without proper assurances/contracts stipulating sufficient security for those resources, right?
If I was writing this, I would probably turn the Summary into a very actionable, short summary and treat most of what's currently there as "Background", or "Justification". Treat the summary from the audience perspective answering "What do I have to do?"
You're talking about why we need something, which is good, but (IMO) that should follow the what we need, and how we would use it.
Thanks!
@mussulma Agreed. I'm just not clear on what the what should be.
added 2 commits
mentioned in commit b7c1a4b4
