Hacker News Reader

bobnamobon E.W.Dijkstra: Simplicity is a great virtue ... complexity sells better.

Unless you've asked for a limit increase _multiple_ times, I can guarantee you haven't reached the limit of step functions.

The default limits are _very_ conservative in large regions

(Admittedly, by the time you've asked for those limit increases you should probably reconsider what you're doing, you're bleeding $$$ at this point)

bobnamobon Fogus: Things and Stuff of 2024

Furthermore, the author mentions ClojureScript earlier in the piece.

I've not done much frontend work, but I'm not sure ClojureScript aficionados would generally feel that TS has much benefit over JS when CLJS is in the picture as well. Salt your JS to your own taste imo

bobnamobon Slow deployment causes meetings (2015)

This is only partly true.

The foundational services do not use cfn for the vast majority of their deployments.

bobnamobon We use our own hardware at Fastmail

nit: Crowdstrike

Unless the incident is now being referred to as “Cloudstrike”, in which case, eww

bobnamobon How we made our AI code review bot stop leaving nitpicky comments

Yet.

The things I've seen juniors try to do because "the AI said so" is staggering. Furthermore, I have little faith that the average junior gets competent enough teaching/mentoring that this attitude won't be a pervasive problem in 5ish years.

Admittedly, maybe I'm getting old and this is just another "darn kids these days" rant

bobnamobon JEP 483: Ahead-of-Time Class Loading and Linking

So you're now weighing the increased container pull time (due to size) vs the class load time you're saving through the cache.

It's nice to at least have the option of making that tradeoff

(And I suspect for plenty of applications, the class cache will be worth more time than (an also probably cached) image pull)

bobnamobon JEP 483: Ahead-of-Time Class Loading and Linking

Prebuilding a cache through a training run will be difficult between lambda invocations though and snapstart[1] already "solves" a lot of the issues a class cache might address.

[1] https://docs.aws.amazon.com/lambda/latest/dg/snapstart.html

Of course, I wouldn't be surprised if the boffins at lambda add some integration between snapstart and class caching once their leadership can get it funded

bobnamobon Time for a code-yellow?: A blunt instrument that works

I think we actually agree with one another. Code freezes/deployment windows should be justified by business risk, not idealised always be pushing mentality.

CDNs and cloud providers are a bit of a special case. AWS definitely has soft blocks during tax peak and large sporting events, CDNs (guessing here) would likely be doing follow the moon(?) (i.e. off-peak) rolling releases

My disagreement with the parent comment was driven by lack of nuance

bobnamobon Time for a code-yellow?: A blunt instrument that works

typo: Reputational not repetitional ofc

bobnamobon Time for a code-yellow?: A blunt instrument that works

Deployment freezes are (should be) completely normal and justifiable.

Do you really think that Amazon is deploying (non-emergency/operational) code changes during BFCM or re:Invent? Or that Fox/CBS/NBC are deploying code changes during the Super Bowl?

Also (unless you have some business need to deploy before the weekend) why deploy on a Friday afternoon? You're just asking for latent defects to page you in over the weekend. Save yourself the trouble and delay that deployment till Monday morning. Your team will thank you in the long run.

Sometimes repetitional and/or sleep damage isn't worth being uncompromising on idyllic CI/CD goals.

bobnamobon Jupyter Notebooks as E2E Tests

There’s also a chance it was placed on the second chance queue and reposted “automatically”

bobnamobon Go Protobuf: The New Opaque API

nit: Coral and smithy are not comparable.

Coral is a schema definition language, yes. But it’s also a full rpc ecosystem.

Smithy at this point is only really an IDL that (in most cases, at least before I left) is “only” used to generate Coral models and then transitively Coral clients and services. The _vast_ majority of Amazon is still on “native” Coral

bobnamobon Principles of Educational Programming Language Design

Omitting an entire class of languages with a strong history in pedagogy damages my opinion of the article.

No mention of Racket, Scheme or SICP!?

bobnamobon Should programming languages be safe or powerful?

grep for .unwrap() and you now have a handy list of every (potential) “null ptr” deref in your rust codebase, try doing the same in Java. You’d need to grep for “.”, I’m sure that’ll be a nice short list

bobnamobon Should programming languages be safe or powerful?

Ime, language “power” is usually used to describe the expressivity and flexibility of the language itself.

As an extreme example, Starlark isn’t a powerful language, it isn’t even Turing complete afaik. This is advertised as a positive attribute for a build description language (I don’t have an opinion, never used it myself).

Conversely, C would classify as a powerful language even considering how barebones it is from a “feature” perspective . Poke any memory however you like is a pretty powerful power

bobnamobon ChatGPT Pro

I wouldn’t bother with Poe, poe2 early access costs $30 and starts on the 6th

bobnamobon DSQL Vignette: Reads and Compute

See yesterdays thread

https://news.ycombinator.com/item?id=42308716

for more

bobnamobon Amazon Aurora DSQL

While I love to take potshots, the yes(configurable) next to that 100gb does some heavy lifting.

I imagine that number is tunable up to multi digit tb if you your monthly bill is worth Jassy's attention

https://docs.aws.amazon.com/aurora-dsql/latest/userguide/CHA...

The more interesting (i.e. constraining) limits are transaction size (both row count and data size) and transaction time, which aren't configurable.

bobnamobon Amazon Aurora DSQL

Marc Brooker's blog has a ~bunch~ bit more technical information

https://brooker.co.za/blog/2024/12/03/aurora-dsql

Specifically, transaction latency is constant relative to transaction statement count, even across region, so I guess that's something

bobnamobon Amazon Aurora DSQL

https://docs.aws.amazon.com/aurora-dsql/latest/userguide/CHA...

The quotas page is always the most interesting for any aws service.

10Mb transaction size and 5min transaction duration limits are interesting limitations that kill any "everything lives in the db" notions that one might have.

A fully transactional, scalable, distributed data store + queue + db + data stream with triggers is my endgame

Also, the 128mb limit on transaction temporary storage seems _very_ limiting for big ol' analytic queries

bobnamobon Amazon Aurora DSQL

The 10MiB transaction size limit smells a lot like FoundationDBs transaction size limit

bobnamobon Amazon Aurora DSQL

Oof, press release proof readers must have been hit by layoffs

> With its innovative active-active distributed architecture, Aurora DSQL is designed for 99.99% availability in single-Region configuration and 99.999% in multi-Region configuration, with an innovative active-active, distributed architecture, making it ideal for building highly-available applications

We get that its architecture is innovative and active-active

bobnamobon Proposed amendment to legal presumption about the reliability of computers

I'm not sure I entirely agree in this circumstance.

Normally I'd give the developers the benefit of the doubt. But the sheer number of issues, and how fundamental some of them ~were~ are[1] leave me little room for sympathy.

https://en.wikipedia.org/wiki/British_Post_Office_scandal#Pr...

Transaction idempotency is such a basic property for a financial system that I struggle to believe that Horizon was tested in any meaningful way.

"the engineers who built the thing" (Gareth Jenkins) are also under investigation for perjury.

[1] Horizon is still in use, in its buggy state, with replacement scheduled for 2030....

bobnamobon Proposed amendment to legal presumption about the reliability of computers

Echoing this, I've been passively following the inquiry but haven't watched the miniseries.

I think it would leave me furious for weeks and push me into a depression rut.

I'm trying to take the whole debacle as long term inspiration to be excellent at what I do - it's tough to stay positive when so many people involved (Fujitsu, PO upper management, the original prosecution) seem entirely morally bereft, with little chance of consequence.

bobnamobon Proposed amendment to legal presumption about the reliability of computers

ye gods, can we come up with a "law" to describe appealing to the halting problem?

Just because there are unanswered questions that doesn't mean we can't have bare minimum codified standards.

Furthermore, standards aren't invalidated just because practitioners disagree with them. Plenty of <insert engineer type>s disagree with the standards body of their respective field, they still follow the standards out of fear of prosecution or simply as a path of least resistance and when those standards are found to be defective, they (generally) evolve.

bobnamobon Proposed amendment to legal presumption about the reliability of computers

> where there was a bug in a system designed by Fujitsu

To clarify, there was not _one_ bug, but hundreds.

https://en.wikipedia.org/wiki/British_Post_Office_scandal#Pr...

From my read of the information publicly available, there was a fundamental lack of distributed/transactional system understanding from the developers at Fujitsu.

This[1] article recently posted here[2] reminded me of the Post Office Scandal.

The lack of respect for established CS theory (transactions and distributed systems) and established accounting practices (double entry book keeping or even the idea of a ledger) is mind boggling.

[1] https://news.alvaroduran.com/p/engineers-do-not-get-to-make-...

[2] https://news.ycombinator.com/item?id=42269227

bobnamobon Jank is now running on LLVM IR

Hi Jeaye, love what you and the other contributors are doing with jank. The further proliferation of clojure dialects can't happen quickly enough imo.

Do you have any "killer app" style use cases in mind for jank? Babashka is great in CLI/FaaS settings, native Clojure is great for "situated" programs that can afford a JVM startup and some memory overhead

Which settings are you particularly excited to use jank in?

bobnamobon Engineers do not get to make startup mistakes when they build ledgers

Kleppmann's Designing Data Intensive Applications is also pure gold.

Worth reading once a year imo

bobnamobon Engineers do not get to make startup mistakes when they build ledgers

For a glimpse at the _real_, human consequences of this kind of slipshod mentality toward money shuffling software see the Post Office scandal[1].

https://en.wikipedia.org/wiki/British_Post_Office_scandal

Anything that moves money should be treated with upmost seriousness and be aware of as many historical mistakes as possible.

bobnamobon Australian Parliament bans social media for under-16s

Counter anecdote, I’ve never been asked for id when voting in South Australia or NSW in my 20 year voting history