397 comments

[ 5.5 ms ] story [ 313 ms ] thread
The best part of this repository is that the authors and the commit dates reflect the true date of the event:

  James Ashley authored and JesseKPhillips committed on Dec 6, 1986

  8th Congress authored and JesseKPhillips committed on Jun 15, 1980
Honestly it seems like one of the worst parts, that the dates are limited by UNIX timestamps and aren't flexible enough to store the dates from centuries ago.
they do when you use signed 64bit timestamps (or 128bit ones if you need the nanos)
Unfortunately git rejects dates before 1970, at least when I set them using GIT_AUTHOR_DATE
This is especially troublesome when you need to cover up those 50 years of procrastination on a project.
You say 'procrastination' I say 'The means IS the end, yo!'.
Apparently it's technically possible in Git as of a year or two ago, with some conversion bugs probably still lurking, but Github/Gitlab still don't support it. And the frontend tools like "git commit" don't support it. The project in this answer is using git hash-object to create the commit from raw bytes. https://stackoverflow.com/questions/21787872/is-it-possible-...
(comment deleted)
No the dates are wrong, and the conversation used is stupid.
So much legislation consists of "add the following text to USC 1.23: 'blah blah blah'" that I'm surprised there isn't something like this for pending bills.
I had this realization like four years ago, and bought `gitvernment.org` precisely to do this...one day.
I should review my purchased domains too
I also had a similar idea for a project that tracked changes so citizens could see exactly which representative changed a bill. The OpenGovFoundation built Madison[0] to enable the public to comment on bills, but that project has been shuttered.

I still think there is some merit to the idea. The challenge, however, is to convince legislative bodies to use such a system.

[0] https://mymadison.io

Agreed, though I think to sharpen the analogy, the United States Code (U.S.C.) would be the "code" in the repository and pending legislation would be pull requests.
There is in a way. One of the selling points of LexisNexis’ law products are these sorts of “diff” views of the law
In my experience these services tend to have terrible UX, they’re not reliable or comprehensive, and they require human maintenance so there are usually huge subscription fees that scale with the number of statutes you have access to. Legislatures really should publish law in a machine readable format so that historical navigation and compilation can be fully automated including by members of the public. The human-readable statues can still be compiled artists; it’s a much easier and lower risk kind of modernisation than, say, converting paper forms into web apps.
ugh, “statutes can still be compiled artifacts”
There's actually a group working on this exact thing at the federal level -- https://congressionaldata.org/bulk-data-task-force-reports-m...

It's a pretty complex task, especially at the state level where there often aren't consistent plaintext publishings, or in some cases even digital versions of the effected code/regulations.

I wonder, how are all the "change this text in this other paragraph" laws carried out then? Wouldn't at least the courts need a "current" version of all laws with all modifications applied?
(IANAL but I do work with digital legal data) — depends on the jurisdiction.

Generally the bills that change legal codes are timed to take effect all in the same near-future date, and so a new “printing” is issued. How often this happens varies, but can be as little as once a year.

In the cases where a law is immediately effected, judges and lawyers are smart and read the news, and services like lexis have alerting. The main legal providers will also “patch” things in themselves in advance of printings, and do change sets and redlines by hand.

Oftentimes bills instruct regulatory agencies to do things, which is a process that includes lots of published notice.

> It's a pretty complex task, especially at the state level where there often aren't consistent plaintext publishings, or in some cases even digital versions of the effected code/regulations.

There are pretty much always digital versions, though they may not be freely available to the public or licenses in a way which would permit using them as a basis. (Lexis/Nexis has them, with change history, links to relevant case law, etc., etc., etc.)

Our nonprofit, Open Law Library, works with governments to do exactly this. For example, the DC Council uses our software to codify and publish their laws in xml and html on github: https://github.com/dccouncil/dc-law-xml. If you're interested in getting your government's laws published using git in human- and computer-readable formats, feel free to reach out to me at dgreisen [at] openlawlib [dot] org.
https://www.legislation.gov.uk/ tracks UK legislation -- you can see "as enacted" and "as amended". They even go through and note where new legislation has an impact on existing legislation without changing the text!

Only for already passed legislation, rather than pending, though.

I love that the commits are attributed to the original authors:

  Amendment XXVII
  Author:    James Madison 
  Date:      May 7 15:00:00 1999 -0400
https://github.com/JesseKPhillips/USA-Constitution/commit/3c...

I wish the commits were dated with the changes' original dates. Perhaps they and GitHub's UI is just confused by pre-1970 epoch dates.

Literally Pre-history
It looks like this is a limitation because Git stores dates as seconds after the Unix epoch (1970-01-01). The underlying file format allows storing negative numbers, which would represent dates before the epoch, but most toolchains don't handle them correctly. Here's a pretty good summary:

https://stackoverflow.com/questions/21787872/is-it-possible-...

I really hope force push is disabled on the repo :)
Nope, a rewrite branch is in the works (to many glaring issues to start with.)
> ...where does the government get the power to ban drugs.

It likes to pretend that the interstate commerce clause gives them that power. _ie_ they're "regulating commerce" by waging the drug war.

See Wickard v. Filburn, whereby the Supreme Court decided that growing crops entirely for your own consumption while staying within the border of one state is still 'interstate commerce'. https://en.wikipedia.org/wiki/Wickard_v._Filburn
Straddle two states with my farm ftw?
I think that will make your situation worse rather than better.
That would technically put almost everything you do that doesn't stay on one side or other, under federal purview. Not recommended unless you're a masochist
This is a buddhistic "everything is connected" legal interpretation.
A better, more modern citation would be Gonzales v. Raich (2005), which upheld federal power to prohibit homegrown cannabis, even in a situation where it was stipulated (IIRC) that Raich's marijuana would never leave her house. https://en.wikipedia.org/wiki/Gonzales_v._Raich

Notably, both Justice Kennedy and Justice Scalia voted in the majority. Chief Justice Rehnquist was as apoplectic about that vote as Kennedy and Scalia were when Chief Justice Roberts sided with the majority to uphold the Obamacare individual mandate.

IMO, the liberal majority had the better argument on the individual mandate precisely because Scalia and Kennedy couldn't bring themselves to draw a line at homegrown cannabis. They were stuck relying on a ridiculous theory about affirmative obligations as they had already blew their chance at cabining the reach of interstate commerce, per se.

People keep posting this case like its supposed to be wrong.

But like the McDonalds Hot Coffee case, when you learn the facts about what actually happened, you realize the case was properly decided all along and those that say otherwise are just spreading ideologically based FUD.

Wickard is not about the crops Filburn grew for personal consumption.

It was about the extra crops he grew just cause. He didn't just grow a little bit extra. He grew hundreds of barrels of wheat extra, meaning several fields more than he was allowed. Filburn claims he was just going to use this excess wheat at home, despite having no means to consume such quantities. Clearly, unless he was actually going to let valuable quantities of crop go to waste, he was intending to sell them to someone.

And if Filburn got away with it, then everyone else would try to do the same. And in a year or two, the national wheat restrictions intended to stabilize the price of wheat would fail as prices collapsed from hoarded inventory entering the market. And wheat farmers would go bankrupt and pull out of growing wheat during a war. So of course SCOTUS ruled against him.

If Filburn hadn't completely blown past the wheat restrictions, there wouldn't be a SCOTUS case about him today.

Shouldn't it have been the state's burden to prove he was going to sell that to someone outside the state?
No. The fact that he harvested way too much was all they needed to prove. He needed to prove it was all for his own private consumption and could not.

Interestingly enough, he won at the trial level on different grounds (challenging the process by which the limit was set) but lost at the appellate level.

(comment deleted)
Disclaimer: not supported within 100 miles of the U.S. border [0], [1].

[0] https://en.wikipedia.org/wiki/Border_search_exception

[1] https://www.aclu.org/other/constitution-100-mile-border-zone

This is great. Is there a tool to do this?

I mean, can I edit the times and the authors and create a history that I need with some software? Or did the author painstakingly go in and manually look up all the dates and make the commits? And then roll back for an error and just keep going?

The issue is how would you change a single commit from 8 commits ago if there was an error there?

And then how would you visualize the whole thing just to keep it straight in your head?

> The issue is how would you change a single commit from 8 commits ago if there was an error there?

Rebase would be my first hunch. There's plenty of GUI based git editors, as long as it's all historically in order it's all fine and dandy.

Rebase and then force-push to reset the remote
My approach would be to generate the git repo using a script of some kind. It wouldn't be particularly difficult, and then fixing a change is just a rebuild.
Would that be a constituting script in this case?
No tools, just manual lookup. There is now a history rewrite based on the feedback happening.
Love this!

And this is the future of legislation that we all need, all bills having detailed attribution to each word and line, linked back to lawmakers and lobbyists who created them!

I wish all legislation would be stored in modern version control.
(comment deleted)
This is probably common knowledge to Americans, but I noticed that the latest amendment on the commit log was written by James Madison. I was confused as to how that came about so recently and found on the wikipedia page:

>It was submitted by the 1st Congress to the states for ratification on September 25, 1789, along with eleven other proposed amendments. While ten of these twelve proposals were ratified in 1791 to become the Bill of Rights, what would become the Twenty-seventh Amendment and the proposed Congressional Apportionment Amendment did not get ratified by enough states for them to also come into force with the first ten amendments.

> The proposed congressional pay amendment was largely forgotten until 1982, when Gregory Watson, a 19-year-old sophomore at the University of Texas at Austin, wrote a paper for a government class in which he claimed that the amendment could still be ratified. A teaching assistant graded the paper a "C" and an appeal to the professor, Sharon Waite, failed, motivating Watson to launch a nationwide campaign to complete its ratification.[1] The amendment eventually became part of the United States Constitution, effective May 5, 1992,[2] completing a record-setting ratification period of 202 years, 7 months, and 10 days

Did his grade eventually get upgraded to A, or was the paper just badly written in spite of this great insight?
The appeal went all the way up to the supreme court. In a controversial 5-4 decision he was awarded a B-.
Did he appeal his grade after that?

  And with the benefit of hindsight, Waite says, Watson clearly doesn't deserve that C she gave him.

  "Goodness, he certainly proved he knew how to work the 
  Constitution and what it meant and how to be politically 
  active," she says. "So, yes, I think he deserves an A 
  after that effort — A-plus!"

  And that's exactly what happened.

  On March 1, Waite signed a form to officially change 
  Watson's grade. Thirty-five years after Gregory Watson 
  wrote his paper, he finally got his C changed to an A.
https://www.npr.org/2017/05/05/526900818/the-bad-grade-that-...
Note well, would-be faculty: this is the degree to which undergrads will hound you if you grade homework accurately.
At least in this case, it wasn't a parent.
Did that end up retroactively changing his grade in the course as well?
Yes, grade changes are done on the course level. The registrar doesn't keep track of the individual assignments.
Papers aren't graded on their real world effects!
> This is probably common knowledge to Americans

I for one never heard of this, so thanks for sharing!

Not common knowledge at all. I learned about it back in the 1970s on a school field trip to Washington DC. I was looking at the original Bill of Rights in the National Archives and realized there were twelve amendments there rather than ten. I pointed this out to my teacher who told me to be quiet. (I think I was 10 years old at the time, so I complied.)
It's probably more commonly known among people who have taken government classes since 1992, since it's basically proof that government classes can have a real impact on public policy.
nope I took US history multiple times after that, still thought the bill of rights had 10 amendments
As far as I know, "bill of rights" is in fact the informal name of the first 10 amendments -- which were proposed and passed as part of the political compromise that got the constitution accepted at all.

The remaining amendments are, well, just amendments. They don't necessarily specify rights, and are generally not considered part of the bill of rights.

Pretty interesting huh? The qualified expert said no way.

A public citizen said yes way and pushed on the government to make it happen.

Back to ignoring the idea so I can push up the stock price.

>When Watson began his campaign in early 1982, he was aware of ratification by only six states and he erroneously believed that Virginia's 1791 approval was the last action taken by the states. He discovered in 1983 that Ohio had approved it in 1873 as a means of protest against the Salary Grab Act and learned in 1984 that Wyoming had done the same 6 years earlier in 1978, as a protest against a 1977 congressional pay raise. Further, Watson did not know, until 1997, well after the amendment's adoption, that Kentucky had ratified the amendment in 1792. Neither did Kentucky lawmakers themselves — in Watson's desire for a 50-state sweep, the Kentucky General Assembly post-ratified the amendment in 1996 (Senate Joint Resolution No. 50), at Watson's request, likewise unaware that the task had already been attended to some 204 years earlier.

>In 2016, Zach Elkins, a professor in the UT Department of Government, became interested in Watson's story and began to document its origins. He tracked down Sharon Waite, who had left academia in the 1980s to work on her family's citrus farm. Elkins suggested to Waite that they change Watson's grade. In 2017, Elkins submitted a grade change form with Waite's signature and a grade change to "A+". In an interview with NPR, Waite stated, "Goodness, he certainly proved he knew how to work the Constitution and what it meant and how to be politically active, [...] So, yes, I think he deserves an A after that effort — A-plus!" The registrar approved a grade change to "A", because the university does not give grades higher than "A".

How has this not yet been made into a comedy?

Would love to see this as a movie. Or at-least a book
>The registrar approved a grade change to "A", because the university does not give grades higher than "A".

It’s hilarious to me that the bureaucracy of this university couldn’t allow even a single grade above an A in this exceptional circumstance - I wonder if it actually came down to some hard limit in their grade keeping system, or if some grumpy recordskeeper just said, “Nope, this is how we’ve always done things, so this is how we will always do things.”

In all seriousness, the profs cant just invent grades that don't exist. Verbal "awesome work from X everybody read it" is for that purpose.

This is not some kind of overreach of bureaucracy.

In all seriousness, the profs can choose to do what they want and the system most certainly can choose to allow it or not.

This is absolutely an example of rigid bureaucracy at work.

Also, you now are eligible for grade changes in past English classes because you used the word "can't" figuratively to describe a literal situation. Whether the grade changes occur and if they go up or down is up to the systems maintaining the grades and the teachers.

Next level grade inflation "A+++", "totally awesome A", "A totally even more awesome then A+++:.

In all seriousness, this is not rigid bureaucracy preventing something that would be useful. Just because prof have such an emotional reaction to someones unusual effort does not mean new grade has a reason to exist.

I guess it depends on the grading system doesn't it? I mean, if the top mark is essentially 100% to mean "there's nothing that reasonably could have been done to improve it" and A is defined as 80-100, then you can't to better than A. Or if the top mark is 1.0 and the worst passing mark is 4.0, and A means 1.0, then you can't do better than A.

They already went way outside, above and beyond anything I could've imagined. How do you change a grade decades after the course? Surely they're locked down soon after. To me, just going off and making it an A (rather than the professor just saying "well I've got egg on my face don't I?") is already incredible and a generous deviation from the bureaucracy.

Oh absolutely! I simply want to challenge the disempowering language of bureaucracy that involves saying someone "can't" do something.

There's literally an infinite number of ways to adapt the system to allow for a higher-than-A grade. Choosing to not do it is still a choice.

I'm not claiming they should've done anything different. I'm merely making sure bureaucracy's responsibility in this matter is pinned down.

"I can't do it" is pretty much normal way how to convey not just it is physically impossible. It normally means:

* It is not up to me to decide that, it would be overreaching my mandate to make that change. * This is against my moral values. * This is against rules that everyone else bothered to follow.

All these are valid choices and right choices. The bureaucracy responsibility in cases like this is literally to resist the pressure to make exception just so someone feels good.

(comment deleted)
The +/- letter grading scale has existed for as long as I can remember. For the particular college I attended, A+ corresponded to a GPA of 4.33 or a % score of >100 and was only given out to exemplary students - I believe this case would certainly be worthy of that extremely high grade.
4.333333333333333333?

Bah! That's grade-inflation, pure and simple. In my day, the grades topped out at cough4.0. Back in my day, a 4.0 coughreally meant somethingcough, not like these cough hold on, I gotta re-up my inhaler-

Ok, where was I?

Oh, right - Get off my lawn!

:)

The case got highest grade available in system that this particular college uses. There is not reason to expand the scale.
Maybe its my European mindset but I find it impossible to believe that any school can just start inventing extra grades.
Maybe it's my southern European mindset, but I find it totally possible.

The world is not as rigid as northern Europe thinks it is...

I'm from the Balkans.
Well, if you're from the Balkans you should surely see it as totally possible!

Far more rule and law-bending is happening than that, for the possibility of "inventing a new grade" to be really impossible/surprising.

Case in point, I know a university case where the professor body wanted to appoint their "own" candidate for a professorship position. The dean and several (majority) professors had already agreed under the table, and the few others didn't matter.

The only problem was they legally had to go through a open contest for the position.

So when someone came up with 5x the qualifications (citations, papers, conferences, better universities attended, etc, globally too) than their ho-hum friend (some relative of the dean), the professor body suddenly folded the position.

A few months later, they announced a new opening, with some specially crafted qualifications, so that only their friend could pass.

I can really see this happening, unfortunately, and I actually did many times. However, Southern bureaucracies are flexible only when they shouldn't be and insurmountably rigid when it's actually good, reasonable, or funny to make an exception. So, no, no extra grade for Watson here. But we might as well create a whole new professional category for the friend of my friend.
New grades are introduced all the time e.g., https://en.wikipedia.org/wiki/General_Certificate_of_Seconda...

> In 1994, the A* grade was added above the grade A, to further differentiate attainment at the very highest end of the qualification. This remained the highest grade available until 2017.

The grades are now 1-9, with 9 being the highest and above A*.

In this particular case, I think it certainly is within a University's remit to award a special grade given the exceptional circumstances.

>The grades are now 1-9, with 9 being the highest and above A*.

Another example of the US inching toward the metric system ;-)

This isn't the US -- these grades are from the UK.

Although yes, the UK too is in this hybrid world where gas (petrol/diesel) is in litres, temperature is in Celsius, but vehicle speedometers are in miles per hour and feet and inches are very popular for height, measuring rooms and furniture, etc.

Don't forget about measuring your own weight in "stones".
They're still miles away from converting.
Seems like they're finding the path.
That's just A levels (which are open to fiddling by the government of the day)

Universities have their own ideas

Probably breaks some column constraints.
Definitely. Adding an extra grade I'm sure would require at least a couple hours of technical work and haunt subsystems and stats analysts for decades to come. Not worth it.
FWIW I remember a (peoplesoft, iirc) roll out that was so painful that certain departments changed their entire course numbering and taxonomy system, because that was easier than teaching the software new tricks.
the school's database may not store strings like "A", "B+", etc. I'd guess it stores grade points, which can be more directly used to calculate GPA. so at schools where A is the highest grade, it is already represented by 4.0. it's probably more trouble than it's worth to accommodate values outside this range just to symbolically correct a grade several decades later.
Yeah that should definitely be part of the commit message!
Yep, this is why git has separate author and commit dates.
This is pretty funny. I had something vaguely similar happen to me, although not nearly as major.

In my matric year (grade 12) I took three first languages (although I already had two first languages by grade 10). The department of education did not like it, but I did it anyway, with some encouragement from my language teachers.

Fast forward to the final exams, I did well enough to make the provincial top 40, but was excluded due to what I still think was a bug in their system. Funnily enough, I walked to the provincial minister for education (later the national Minister of Health) and told him what happened. For a short while I was a radio celebrity due to the minister personally telling people "Look, this white boy got a B in Sepedi so you guys really should stop complaining and do better." [1]

The error was not corrected and I was actually at the ceremony by chance for having gotten best in English in the province. I also randomly got a call from some trust and got about $100 for my creative writing essay.

In any case I guess the whole radio thing was more amusing to me and actually was a better prize in the end.

[1] In South Africa, a B is 70%—79%; I had gotten 76%.

Of note, there is another amendment in a somewhat similar position, the Equal Rights Amendment: “Equality of rights under the law shall not be denied or abridged by the United States or by any state on account of sex.”

It's one state shy of passing and has been that way since the '80s. That said, there's a very strong argument that the amendment is no longer eligible to be passed because, unlike Madison's amendment, there was an explicit "this offer expires" date in the bill passed by Congress, and that date has long passed. Also, several states rescinded their ratifications. That said, there are plausible legal theories that states aren't actually allowed to un-ratify amendments and that proposed amendments can't expire.

>Keep in mind this is not a formal platform to establish Constitutional changes.
Yes, I loved that comment!
Love this! I just wish the Bill of Rights had been a pull request that was merged once the 11 ratifying states approved it.
This is super cool. I wonder if it could be taken a step further?

Because to me, the interesting parts are not the final versions, but would be in seeing the version history of the constituion as it was drafted and language was changed and rechanged, even with commit notes taken from records of the constitutional convention.

For example, the currently topical impeachment language went through revisions on "be removable on impeachment and conviction of": [1]

> "malpractice or neglect of duty." -- two North Carolina members

> "treason, bribery, or corruption." -- five-member drafting committee

> "treason, bribery, or maladministration." -- George Mason

> "high crimes and misdemeanors against the United States." -- delegates

> "high crimes and misdemeanors." -- final version

It really would be fascinating to watch the constitution evolve in "real time" across all edits. I don't know if anyone's done something like that before. Obviously it would be a tremendous amount of scholarly work to put together.

[1] https://www.latimes.com/archives/la-xpm-1998-dec-19-mn-55593...

The commit messages would eventually all be WIP or somedoc_version_xx.
I think this should be reviewer comments on the pull request or issue that drove the change. It's not nicely representable in git, but it is on GitHub.
Can I submit a pull request to get rid of the phony bits? Perhaps I'd better make that a series of pull requests.
I'll bite...what are examples of phony bits?
This was a reference to news stories of the past few days characterizing parts of the constitution as "phony".

The people who downvoted my comment were probably doing the right thing: I think the comment is funny and germane in the sense that git is a revision control system, but the comment can be inflammatory and really only marginally adds to the discussion.

For the record I can't characterize any ratified part of the US constitution (even parts no longer applicable and/or deleted, such as the prohibition amendment #18) as "phony" (whether I think they are a good idea or not) as they are, by definition, authentic.

This is very cool. It would be nice to add commits for all the drafts of each amendment. For example, here are various drafts of the 2nd: https://en.wikiquote.org/wiki/Second_Amendment_to_the_United... It used to contain the phrase "but no person religiously scrupulous of bearing arms shall be compelled to render military service in person" but this phrase was dropped in the final version.
What I find incredible is that the LEGAL industry has yet to adopt any 'git' style repo to manage all the multi-touch contract edits.

My gut tells me the simple reason is because that would effectively make them lose most of their arcane billing hourly fees.

Many lawyers don’t work by the hour though, so what explains them?
They bill in 6 minute increments, right?
I was referring rather to corporate lawyers who are paid salaries rather than hourly rates.
Word documents have extensive markup and revision tracking features, and then they're passed around within emails that function similarly to commits. Git is only really useful when you're doing simultaneous changes to a large number of files.
In my experience Word's revision tracking is entirely inadequate to the task. I might take that back if everyone was editing a shared Word doc on Office365, but the typical scenario of emailing versions back and forth makes it very difficult to keep track of the current state of the document and ensure everyone is looking at the same thing.

I don't think standard git with MS Word blobs would be better though.

What I would like would be the conceptual structure of a git version history but with a commit viewer that could represent the commits in Word's revision control style as well as pull requests to track incremental agreements to change the document.

But would version control solve the email issue? I guess I would still rather get an email with a git repo or patch than I would a word document with a revision set.
JesseKPhillips was my favourite founding father. He was a genius for adding whitespace and readme.
> Add whitespace for better diffs

For those confused..

There is a commit saying this

I kinda wish they'd rebase the repository to put that stuff in so git blame would still give you the original authors.
As I've said before, the Constitution and our current legislative models were devised in a time of very slow travel and communications, when it made sense to hold periodic elections and have legislators travel to a central location for terms.

There are no technical barriers to putting our entire legislative code online, dispensing with most legislative organs, and allowing people to edit it directly - a Wikiocracy, so to speak.

Although this sounds reasonable having everything published and voted on in real time turns out to just end back up with a representative democracy. This article explains some of the challenges with that idea much better than I can. https://medium.com/civic-tech-thoughts-from-joshdata/so-you-...
This isn't what I'm proposing at all. I do get his points, which are sincerely made, but I have also been thinking about this for decades and put in a lot of time in participation and advocacy at all levels of the political process.

I'm proposing getting rid of legislatures and the whole legislative process and having the legal code publicly accessible for editing by the public. It would be on a blockchain (for verifiability purposes), and there would a derivative graph (for functional evaluation purposes - quite necessary because the whole meaning of a law could be changed by a relatively tiny textual edit like 'you must do X' -> 'you must not do X').

Up to now most network analysis of law has been applied to the legal infrastructure that grows around individual cases and treating legal code itself as axiomatic; consider, rather, analysis of the code's own network properties as tremulously explored in section VI herein: http://digitalcommons.law.umaryland.edu/cgi/viewcontent.cgi?...

The most immediate benefit of fully graphing the constitution/code/cases ecosystem would be the possibility of a serious pruning, to address the problem first pointed out in The Republic of a surfeit of laws that grows to such complexity that it crowds out confident understanding by the citizenry.

We are quite a long way from replacing our existing legislative technology with a publically edited version, but the time to start exploring this is now, given the increasingly precarious state of our institutions and the widening asymmetry between our technological and legislative capacities. Put another way, we are innovating aster than our legal or political systems can keep up, and absent a consensus or voluntary dissensus on the social deployment of such innovation, it is more likely to be weaponized than any other outcome.

Based on our experience with Wikipedia, that would be the worst idea imaginable from a social and political perspective.

The wiki would just become a series of editing wars with multiple parties attempting to be the last to edit the wiki. For any sort of sanity, you'd need to lock down most of the wiki, putting the power in the hands of a few editors. And then you'd have the problem of selecting the editors....which would bring you right back to where we are today.

Not at all. You'd just have to get buy-in for commit changes. There is no editorial selection, the whole point of wikipedia is that anyone can edit it. I am not arguing that Wikipedia as-is is ready for deployment as a legislative corpus, but that no technical barrier exists to the creation of something built along similar lines.
There are significant technical barriers to what you're describing.

For starters, how do you limit edits to citizens? How do you verify citizenship? How do you handle racing commits? How do you determine "buy-in"? How do you store the data securely? How do you secure the authoritative single repository of all law for your jurisdiction? If distributed, how do you determine the authoritative source? How do you trust change commits coming from external sources such as the other distributed stores of the liki?

Not really. First, I don't, only to residents. With brakes, much as W handles edit wars. Buy-in is most likely a function of affective scope, and itself is up for debate. On LOC.gov, eventually or using some sort of blockchain.

I don't think most of these objections are technical so much as social. Nor am I offering it as a complete recipe ready to go; it's a proposal with very manageable problems, certainly no worse than those Wikipedia faced when first instantiated. You have to start somewhere.

Are the amendments really meant to be inserted at unspecified points in the main text? I never thought of them like this, and it doesn't always work well here.

Most glaringly, there is the part in https://github.com/JesseKPhillips/USA-Constitution/commit/7f... in the article talking about the President's powers, saying "The President shall have Power to [...] He shall from time to time [...]" and the diff inserts "No soldier shall, in time of peace be quartered in any house, without the consent of the owner, nor in time of war, but in a manner to be prescribed by law." between these sentences.

This doesn't work at all, not only because this amendment has nothing to do with the President's powers, but especially because the "He" starting the next sentence has nothing more to refer to.

Of course this would be a lot more boring if the amendments were just tucked on at the end, but I'm afraid reality is just boring here.

> Are the amendments really meant to be inserted at unspecified points in the main text?

The US constitution is unusual in leaving the main text unchanged and appending amendments at the end. Most countries don't do that.

For example, Australia's constitution [1] has been amended several times since it was first adopted in 1900. But, it is amended by changing the text of existing sections, inserting new sections or subsections (examples: section 105A, subsection 51(xxiiiA)), and deleting entire sections (example: section 127).

[1] https://www.aph.gov.au/About_Parliament/Senate/Powers_practi...

Most notably the 21st Amendment repealed the 18th, but both texts are still present in the amendments section of the US Constitution.
This is pretty funny. I had something vaguely similar happen to me, although not nearly as major.

In my matric year (grade 12) I took three first languages (although I already had two first languages by grade 10). The department of education did not like it, but I did it anyway, with some encouragement from my language teachers.

Fast forward to the final exams, I did well enough to make the provincial top 40, but was excluded due to what I still think was a bug in their system. Funnily enough, I walked to the provincial minister for education (later the national Minister of Health) and told him what happened. For a short while I was a radio celebrity due to the minister personally telling people "Look, this white boy got a B in Sepedi so you guys really should stop complaining and do better." [1]

The error was not corrected and I was actually at the ceremony by chance for having gotten best in English in the province. I also randomly got a call from some trust and got about $100 for my creative writing essay.

In any case I guess the whole radio thing was more amusing to me and actually was a better prize in the end.

[1] In South Africa, a B is 70%—79%; I had gotten 76%.

Edit: This is meant to be a comment on the latest commit—the 200 year old ammendment that got ratified in 1992, essentially due to a C being awarded for an essay about how the 200 year old ammendment could be ratified.
Note: this is now a duplicate, happy if mods remove this.
Man I wish ever bill and law was under version control so you could do a git blame
There is rarely a single person behind changes though. You would have to show a list showing the person who proposed it, the people who requested changes to it and the people who voted for it.
The actual writing of the words is usually outsourced.

Quite a few people are involved in that. Clerks, lobbyists, lawyers, etc.

Even the reading of the laws that lawmakers vote on is outsourced. (Ie lawmakers often don't read what they vote on.)

That is why you have merge request approvers and the merger. The work can be authored by many because each makes small contributions.

You can even have multiple protected branches where "policy makers" merge the work from their lawyer.