Open masinter opened 7 months ago
Here are some questions and notes to help understand the problem.
What's the scope of the plan? Does it encompass all publications, including material of historical interest, or only some defined categories such as manuals and technical documentation?
Should interlisp.org URLs provide a single and stable namespace for all documentation and publications? For example, interlisp.org URLs may directly serve or redirect to the linked documents stored at interlisp.org or elsewhere.
Should all the material be stored at a single place like GitHub? Other storage areas may provide desirable features such as access control for material that can't be shared (e.g. Google Drive or the CHM PARC archives).
Where does Zotero fit in? Should it be only a bibliographic database with links to the sources or also store part of the material?
What's the scope of the plan? Does it encompass all publications, including material of historical interest, or only some defined categories such as manuals and technical documentation?
Everything we have has to be somewhere, or else we don't "have" it. There may be some things on the net we can trust to keep things, but best if we 'have' copies something is in more than one place, then one of the places should be designated as the 'home' and other places are either historic or distribution copies.
Should interlisp.org URLs provide a single and stable namespace for all documentation and publications? For example, interlisp.org URLs may directly serve or redirect to the linked documents stored at interlisp.org or elsewhere.
I think 'single' and 'stable' are aspirations. A web server is a portal, a gateway. We should try to avoid having more than one URL for the "same" resource.
Should all the material be stored at a single place like GitHub? Other storage areas may provide desirable features such as access control for material that can't be shared (e.g. Google Drive or the CHM PARC archives).
I don't think it's practical.
Where does Zotero fit in? Should it be only a bibliographic database with links to the sources or also store part of the material?
Zotero seems to be a tool for a single person -- without version management I want to look for a different tool.
Here's a view of the current Interlisp.org site. We can use this as a base and make modifications and additions using this as our starting point.
graph LR
Interlisp.org --> history
Interlisp.org --> project
Interlisp.org --> software
Interlisp.org --> documentation
documentation --> html
history --> bibliography
history --> d-machines
history --> glossary
history --> guis
history --> lisp
history --> medley
history --> in1([in-memoriam])
history --> parc-alto
history --> timeline
project --> comments
project --> getInvolved
project --> partners
partners --> p1([SHFT])
partners --> p2([SPN])
partners --> p3([educopia])
partners --> p4([nterlisporg-inc])
project --> status
status--> s1([2021MedleyAnnualReport])
status --> s2([2022MedleyAnnualReport])
status --> s3([2023MedleyAnnualReprot])
project --> pro1([credits])
project --> pro2([faqs])
project --> pro3([organization])
project --> pro4([reviving])
project --> pro5([stories])
software --> access-online
software --> install-and-run
software --> using-medley
install-and-run --> linux
linux --> linux_images[images]
linux --> l1([linux-local-from-github])
linux --> l2([linux-standard-from-github])
install-and-run --> macos
macos --> mac_images[images]
macos --> m1([macos-from-github])
install-and-run --> windows
windows --> win_images[images]
windows --> native
native --> n1([widnows-native-from-github])
windows --> wsl
wsl --> wsl1([windows-wsl-local-from-github])
wsl --> wsl2([windows-wsl-standard-from-github])
using-medley --> use1([d-using])
using-medley --> use2([il-using])
using-medley --> use3([keystrokes])
using-medley --> use4([online-using])
It's a good start, but the scope is broader
For "where this are" and "where things should be", there are several places: Google Drive -> The Interlisp google drive, the CHM-parc drive YouTube GitHub: Various repositories and access paths to them Various other data in GitHub (issues, discussions, etc.)
Key limitations: we don't want to grow the medley repo to hold videos or even PDFs (especially not PDFs that have scanned images). Google Drive has a limit on the number of files in a single drive.
Zotero has a disk file limit.
But if you look at https://interlisp.org/software/using-medley/ there are some things listed as "Adanced Material" and "Unsorted documentation content". What are those? Do they fit into categories? What are the requirements for users to find what they are looking for?
Can we separate out the historical references to documentation for older versions (which we should leave alone except perhaps to annotate or reformat to provide better access) and the documentation for current Medley, which we should be updating. There are meeting recordings and summaries that are scattered about, between Zoom and Google Drive and Fathom and YouTube. There are PARC videos about Interlisp at Internet Archive.
We need to agree on a set of rules for what files go where, that is comprehensive and meets all of the requirements. I'm not sure what all of the requirements are, but
Comprehensive: It should include: what files are in the repo, in the google drive, in the "export from repo as PDF" files, those that we obtained from the PARC CHM folders, what things are indexed in Zotero, etc.