Pete Davidson joked that depression is a rich person's condition because it implies you have a life you shouldn't be depressed about. Same with software engineers?
1,169 karma · joined December 16, 2010
Let's say you're working on dev, you add two commits, and now you want to merge these to main.
Most people don't know that GitHub has no PR merge method that simply allows you to copy those two commits onto main.
It's literally impossible for you to use GitHub PRs to simply have dev become main.
Here's why:
If you use a merge commit, it will create a merge commit ("The pull request is merged using the --no-ff option.") So dev isn't becoming main; main is becoming something new.
If you use squash and merge, it will rewrite your two commits into a single commit. But you wanted two commits or you would have already done one commit in dev. Again, dev isn't becoming main; main is becoming something new.
If you use rebase and merge, it should then copy your two commits onto main, but in fact, it rewrites them with new commit hashes, it's not a copy. Then, you need to rebase dev back onto main because main is now something new.
More detail: https://github.com/orgs/community/discussions/4618#discussio...
I'm a total amateur here.
Is there no hope of finding a new source of helium?
Is there any opportunity for airship gasses that are reasonably competitive to helium but have a much more abundant supply? Is helium the only possible such gas (besides hydrogen)? Are competing gases 30% less efficient or 3,000% less?
A lot of crypto people knew he was unscrupulous and opposed him before FTX blew up. There were a lot of signs he was a bad person and the effective altruism thing was B.S.
But, pretty much nobody in crypto expected him to be the wholesale incompetent fraud he ended up being. We thought he was a bad guy, not a blood-chillingly evil moron.
It's nice to see the NYT shift to offer some reasonable coverage on Sam. One of the NYT reporters, David Yaffe-Bellany, is a good friend of Sam, spoke with him regularly even after he was apprehended, and routinely provided rose-colored puff pieces throughout his fall.
If the CEO of a factory had to sell half their maintenance-hungry machines to balance the budget, do they owe it to the sold or retained machines to take a pay cut so they "share in the harm of the layoffs"?
The factory machine example is silly because they are machines and offensive because many people would agree that employees are very much not machines.
In pretty much every major city in North America, it is illegal or there is too much red tape to build the affordable housing people actually want.
Why is it that living in a tent on the street costs $0/month but the entry-level apartments cost ~$2k/month? Why are there no price points between $0 and $2k? The answer is bad government policy.
At the time of the merge, the most economically productive use for graphics cards was to mine Ethereum. The merge caused a worldwide crash in graphics cards prices. GPT 3.5 launched two months later.
If Ethereum had not merged and instead stayed on proof-of-work, it would have consumed 0.2% of global electricity over the past 365 days.
The merge was completed with zero network downtime, like swapping the engine of an aircraft mid-flight.
For example, using Ethereum L2s or other public chains, anybody in the world with a phone will be able to access American T-Bills and equities, lending products, and much more.
Public chains are a better foundation for a global free market economy, and that includes but is not limited to payments.
You're a long-term hacker news community member. Like all of us, you pride yourself on the rational exploration of technology. Does "They've had more than a decade to find a single usecase other than crime, and they haven't" strike you as a particularly rigorous argument? Why would a decade be the magic number? Might there be more objective or intrinsic metrics of realization of potential than a set number of years? For example, Ethereum invented the crypto app layer and launched eight years ago, so perhaps ten years is both arbitrary and incorrect?
You seem to be somebody who is certain that crypto is- paraphrasing- useless for anything but crime. Given your conviction in this view, what up-to-date evidence do you have? For example, how would you contrast the capabilities of modern non-blockchain payment systems, like FedNow or India's UPI, versus the emerging network of public blockchains? If you can't answer the question, maybe your strong conviction is unjustified?
It seems that many technologists in the HN community have been whipped into an anti-crypto frenzy by crypto's many scams and now, blinded by rage and disgust, are unable to pursue any semblance of the kind of rational inquiry that otherwise is a fixture in their intellectual lives.
As far as I know, it's impossible to merge a GitHub pull request without one of squashing, generating a merge commit, or rewriting the commits. It is not possible to simply copy pre-existing commit IDs.
You can test this yourself. Simply checkout the main branch, add two new commits, push those commits to a dev branch, and try to merge them in a PR to main in a Rebase: you will observe that github has rewritten the commit IDs vs. the ones you generated locally on commit.
Today, if you want to preserve commits in your development branch (and not squash them; squashing is great sometimes), you have two bad options:
1. merge commits. This is bad because linear history is an excellent risk reduction mechanism for a production build.
2. rebase commits. These would be ok, except when github rebases the commits into linear history, the commit hashes change (they are new commits) even though the patch and metadata are identical. So your development branch itself constantly needs to be rebased on your production branch, because the commits being added to production are newly created by github rebase & merge.
What we need is a fourth option "Fast-forward merge", which requires linear history and only preserves (ie copies) commits from the PR source branch into the target branch.
How can we get GitHub to care about this and action on it?
Crypto bro response: loaning my USDC on Notional for a fixed rate of 4.6% and then bridging it to Arbitrum Nova to then send my friends and family interest-bearing US dollar payments for zero transaction fees seems pretty f'ing useful to me.
Degen response: lol ok bro gl with that
In other words, both USDC and uninsured SVB deposits had the same market response over the weekend. But every USDC holder had unfettered access, whereas only large SVB depositors were able to access the informal market of private sellers.
I want you to remember this interaction because by the end of the decade, there will be over $1 trillion of USD stablecoins in circulation (in today's dollars), and at least one of the five largest banks in America will run a major stablecoin product of some kind, such as their own stablecoin, redemptions for an existing coin, or consumer on/off-ramps to the crypto financial system.
But a lot of middle-class people try and fail to create any kind of business.
The reason is that it takes more than money and education to build a business.
Money and education are necessary to build a great business, like electricity is for your home, but insufficient, as your home also needs good construction, running water, etc.
Elon's inherited wealth and excellent upbringing were necessary for his Tesla and SpaceX success, but far from sufficient.
If a good rich childhood was all that's necessary to commercialize electric cars, reusable rockets, and fast satellite internet, then we'd have a lot more of these kinds of amazing products as millions of people have good rich childhoods.
Yet, non-custodial holding is becoming much easier and safer.
Crucially, wallets are improving a lot, and a spectrum of custodial options is developing, with rich tradeoffs in eg. safety and self-sovereignty.
For example https://www.coinbase.com/blog/how-smart-cryptography-makes-c...
However note
- SVB was closed on a Friday (as is the FDIC's custom). This meant that USDC could not process redemptions over the weekend as banks were closed, and this created fear in the market.
- USDC had 8% of their reserves trapped in SVB. The fair market value of USDC would have been $0.92 if all SVB deposits were lost, which was never likely.
- crucially, USDC operator Circle has acknowledged the need to transition to a "full reserve" model where reserves aren't subject to bank runs. https://twitter.com/jerallaire/status/1635548066185871360
- over the weekend, USDC did lose its $1 peg and trade as low as ~$0.88. However, all USDC holders were able to maintain access their USDC to do whatever they wanted with it, which is a lot better than SVB depositors frozen and waiting to see what happened.
However, American-run USDC is growing faster and has excellent transparency:
In the near future, stablecoins like USDC will become immune to bank runs because the US Dollar reserves backing them will be held in vehicles that don't loan out the reserves and hold short-duration treasuries directly with the Treasury Department.
At the same time, any person or entity with USDC in their wallet has instant, 24/7 access to global markets at their sole discretion, without intermediaries.
On Ethereum today, anyone can buy Coinbase stock, treasuries, an S&P 500 index, and real estate.
In short, the UX of stablecoins is becoming vastly superior to bank deposits because you'll be immune to bank runs, control your own money, and have instant access to global markets, including for low-risk yield on your stablecoins, such as in treasuries or over-collateralized lending.