tneely

I've always thought it would be interesting to apply this sort of approach to health data. I own my data and control where it's stored and how it's accessed. And hospitals anywhere can interact with my data as needed while labs, doctors notes, etc. all live with the hospitals.

DSQL uses an internal service, Journal[1], for its mutli-region writes.

[1] https://www.allthingsdistributed.com/2025/05/just-make-it-sc...

I’m really interested to hear about the technical changes that needed to happen to make this work.

I took a similar approach with LaTeX, where I have a GitHub workflow [0] that regenerates my resume on commit.

[0] https://github.com/tneely/resume/blob/main/.github/workflows...

Yeah that's fair - it doesn't need to be cryptographic. But someone back in the day decided MD5 was what they wanted and it stuck. It always raises alarms with pen testers and security scans at work, and each time we have to explain that the cryptographic security is irrelevant; it's just some unfortunate genomics standard we need to support.

We still see heavy use of MD5 in genomics as well. It's effectively used to generate a single identifier that can be used to reference a specific genome assembly. There have been discussions and attempts to move to other, more secure algorithms, but the community and its tooling is too deeply entrenched in using the MD5 for the reference that it would take a herculean effort to change.

I'm personally of the opinion that it doesn't matter. MD5 is fine for genomics. The chances of valid genome files colliding is still extremely low, and there's not really any relevant attack space. Replacing one assembly file with another will just break someone's analysis pipeline, and most likely in a very clear obvious way.

I'm quite curious about this too - both from a cost and performance perspective. If S3 Express is close enough to EFS on these metrics, then I'd say it wins out due to the sheer ubiquity and portability of S3 these days.

I’ve effectively worked on greenfield projects my entire time at Amazon:

1. New service within Prime to handle GDPR and other compliance related matters

2. Opensource CLI tooling

3. New AWS service

4. Another new AWS service

Maybe (1) doesn’t count since it had to operate within a preexisting microservice ecosystem, but in the rest we’ve had complete control of the product from languages to servers to cloud infrastructure.

I’d wager there are always new teams in most large companies that are doing greenfield projects, you just have to look for them and be willing to join.

I feel like this should be inverted: make friends with those who live near you. I understand we can’t be friends with everyone, but chances are there’s 3-4 people in your immediate area that you’d get along great with. I think it’s a failing of community that we aren’t closer to our neighbors. Sure living close to high school / university friends would be great, but as many posters have noted, circumstance isn’t always that convenient. We need more events that promote socialization amongst neighbors in the hopes that some of those interactions will result in meaningful friendships.

While I agree it doesn’t matter much 1-2 jobs into your career, Ivies are really good at getting you through that first door. I wouldn’t be surprised if that MIT degree got someone to their first technical screen at Google, or gave them the right resources they needed to pass the interview.

I like to start with Skeleton CSS or similar, strip out all the column stuff, and just rely on grid and flexbox to build the layout.

Ruling can be found here: https://www.supremecourt.gov/opinions/17pdf/16-285_q8l1.pdf

It is generally accepted that long-term memory (LTM) is encoded as alterations in synaptic strength.

RNA from a trained animal might be capable of producing learning-like behavioral change in an untrained animal.

Why did the author jump to the conclusion that RNA == LTM, when that RNA is most likely just the driver for modulating synaptic strength? Whatever RNA they extracted could easily just encode for various synaptic proteins.

Facebook, maybe. Social media? No way. I feel the connections it brings us, or helps us maintain, are too important to lose. If Facebook can't do it right, someone else will step up to fill the gap. Hopefully with a less malicious business model.

It's updated now.

What can be done to mitigate the risk of handoffs? This seems like something we can use technology to solve.