News for hackers. Stuff that matters.

elvinyung

Re:Behind the scenes at the NeXT Computer office on Monday October 26, 2026 @09:38AM (#50243470)
which andy are you? :p
Re:Carta to exit secondary trading business on Tuesday January 09, 2024 @01:47AM (#38923165)
How was the beef stew?
Re:A software change allowed FTX to use client money on Wednesday December 14, 2022 @01:15AM (#33980585)
https://calpaterson.com/bank-python.html

> One of things that tends to boggle programmer brains is while most software dealing with money uses multiple-precision numbers to make sure the pennies are accurate, financial modelling uses floats instead. This is because clients generally do not ring up about pennies.

Re:The hunt for a cluster-killer Erlang bug (2021) on Tuesday June 14, 2022 @09:14PM (#31748029)
That doesn't seem mutually exclusive with the fact that the oldest service is written in the oldest-used language.
Re:Arcology: The city in the image of man (1969) on Thursday August 05, 2021 @02:43PM (#28077809)
I learned about Paolo Soleri from William Gibson's Count Zero (sequel to the much more well-known Neuromancer). Gibson always had a knack for foreseeing not just the automobile, but also the traffic jam:

> But today’s episode kept veering weirdly away from Michele’s frantically complex romantic entanglements, which Bobby had anyway never bothered to keep track of, and jerking itself into detailed socioarchitectural descriptions of Soleri-style mincome arcologies. Some of the detail, even to Bobby, seemed suspect; he doubted, for instance, that there really were entire levels devoted to the sale of ice-blue shaved-velour lounge suites with diamond-buckled knees, or that there were other levels, perpetually dark, inhabited exclusively by starving babies. This last, he seemed to recall, had been an article of faith to Marsha, who regarded the Projects with superstitious horror, as though they were some looming vertical hell to which she might one day be forced to ascend.

> Other segments of the jack-dream reminded him of the Knowledge channel Sense/Net piped in free with every stim subscription; there were elaborate animated diagrams of the Projects’ interior structure, and droning lectures in voice-over on the life-styles of various types of residents. These, when he was able to focus on them, seemed even less convincing than the flashes of ice-blue velour and feral babies creeping silently through the dark. He watched a cheerful young mother slice pizza with a huge industrial waterknife in the kitchen corner of a spotless one-room. An entire wall opened onto a shallow balcony and a rectangle of cartoon-blue sky.

Re:Square to acquire Afterpay for $29B on Monday August 02, 2021 @02:21AM (#28033452)
Would you say the same about a house? A car?

You might say that buying a house on credit isn't frivolous but a pair of shoes on credit is, but where do you draw the line in the middle?

(Disclaimer: I work at Square, but I am otherwise completely uninvolved with this acquisition or product area.)

Re:Experts Doubt the Sun Is Burning Coal (1863) on Tuesday May 05, 2020 @01:50AM (#23076916)
Maybe "he" is short for "Helios" :-)
Re:Magic Leap trying to sell for $10B on Wednesday March 11, 2020 @10:56PM (#22553221)
I guess the word "magic" is cursed. General Magic, GetMagic.com, and now this (not that we didn't see it coming)...
Re:Why Go Was the Right Choice for CockroachDB (2015) on Monday February 03, 2020 @02:39PM (#22228096)
Needs a (2015).
Note that as you mentioned since most shops in Japan specialize in a specific type of soup (i.e. the only differentiator is the tare and the toppings), it might be more useful to familiarize oneself with the characteristics some of the more common regional styles. For example, "Hakata-style" ramen generally describes a creamy tonkotsu soup, "Kitakata-style" is generally a chintan with shoyu tare and hand kneaded "temomi" noodles, Yokohama "iekei" generally tends to be a paitan with shoyu tare, Sapporo ramen is generally a pork soup with miso tare, etc.

This is not to say that e.g. Sapporo only has Sapporo-style ramen or Fukuoka only has Hakata-style ramen, but the names describe where the style is perceived to have originated from.

The water is slightly muddied (no pun intended...) by techniques like "double soups", i.e. mixing multiple soups in the same bowl. In some cases a chintan might be mixed with a paitan, etc.

Re:How William Gibson Keeps His Science Fiction Real on Monday December 09, 2019 @02:38PM (#21745656)
Remember all that stuff about the Cayce Pollard Units :P
Re:What Is the Cost of Building a Subway? [pdf] on Thursday December 05, 2019 @06:21PM (#21717761)
I also like this analysis of general high infrastructure costs in America: https://pedestrianobservations.com/2019/03/03/why-american-c...

This tidbit is particularly infuriating:

> In California, the problem is, in two words, Tutor-Perini. This contractor underbids and then does shoddy work requiring change orders, litigated to the maximum. Ron Tutor’s dishonesty is well-known and goes back decades: in 1992 Los Angeles’s then-mayor Tom Bradley called him the change order king. And yet, he keeps getting contracts, all of which have large cost overruns, going over the amount the state or city would have paid had it awarded the contract to the second lowest bidder. In San Francisco, cost overrun battles involving Tutor-Perini led to a 40% cost overrun. This process repeated for high-speed rail: Tutor submitted lowest but technically worst bid, got the contract as price was weighted too high, and then demanded expensive changes. It speaks to California’s poor oversight of contractors that Tutor remains a contractor in good standing and has not been prosecuted for fraud.

Edit: oh, wait, just realized this is from the same blog, so the same body of work.

Yes -- in fact, the original Spark implementation ran on Mesos only, as per the original RDD paper by Zaharia et al.: https://www.usenix.org/system/files/conference/nsdi12/nsdi12...
Re:Viral Tweet About Apple Card Leads to Goldman Sachs Probe on Saturday November 09, 2019 @05:50PM (#21495038)
I mean, don't you think it's kind of a strawman to assume anyone thinks machines are explicitly sexist? It seems much more likely that whatever method they use to do feature selection on this obviously high-dimensional data just happens to end up picking something that was a proxy for gender.

You can be implicitly biased without being aware of it. This is true for both humans and algorithms.

Re:Airbnb Plans to Verify Every Listing on Wednesday November 06, 2019 @03:10PM (#21466629)
In case anyone missed the context, this is probably related to https://www.vice.com/en_us/article/43k7z3/nationwide-fake-ho...
Re:Books: an immutable double-entry accounting database service on Thursday October 17, 2019 @06:59PM (#21287138)
Yep :P

For a fun fact, CockroachDB was actually started by ex-Squares, so we're definitely very familiar with this exact same argument internally, back in the day when neither CockroachDB nor Spanner existed as a viable option :)

Re:Books: an immutable double-entry accounting database service on Thursday October 17, 2019 @05:34AM (#21280280)
Right -- that's why I said most RDBMS. To get ACID and CAP-consistent (which, as an aside, are not the same thing) cross-shard transactions requires some form of migrating your production system onto such a distributed-capable RDBMS that you describe (of which, Spanner arguably fits the description of).
Re:Books: an immutable double-entry accounting database service on Thursday October 17, 2019 @03:53AM (#21279836)
(Disclaimer: I work at Square, but not on Books.)

>Preventing inconsistencies by enforcing constraints is a key point of an RDBMS. This is literally why people use SQL DBs.

This comment is not specific or related to the work at hand, but note that this is only true for a single-machine database or a sharded setup in which you never have to perform transactions across shards. That is, the ACID guarantees are only enforceable in most RDBMS if all your data is on the same machine.

Just curious, do you have a good idea of how many of these are still being actively developed? I remember seeing stuff like Ringpop and Hyperbahn get moved to the uber-archive github account.

I'm especially curious about Schemaless, since it seemed like at one point every company tried to build a NoSQL KV store on top of sharded MySQL, realized how janky it is, and are currently decommissioning it.

Re:Michel Foucault: Power and Struggle on Thursday October 03, 2019 @05:51PM (#21152622)
Foucault doesn't use the word "power" the way you think he does.
Re:Leadership Transition Announced for MIT Media Lab on Wednesday September 11, 2019 @04:08AM (#20937643)
> "Tasting Menu in Zero G"

Meh. The Media Lab is (or was) doing some of the best work in the field in food and flavor science. Arielle Johnson [1] in particular is known for her work and expertise on flavor and fermentation, and she went on to lead R&D at Noma's science bunker [2] and now has her own lab in East Village [3]. She's probably has had an outsized impact on the food world in the last decade or so on stuff like fermentation, which is gradually making its way beyond just fine dining and is something that I'm particularly excited about.

[1] https://www.ariellejohnson.com

[2] https://www.eater.com/2018/3/9/17100494/rip-noma-science-bun...

[3] https://www.ediblemanhattan.com/magazine/summer-2019-issue-n...

Re:Waltz: A Distributed Write-Ahead Log on Monday September 09, 2019 @06:47PM (#20923347)
Just curious, have you looked into Redis streams?
Re:Waltz: A Distributed Write-Ahead Log on Monday September 09, 2019 @06:38PM (#20923260)
Yeah, I'm currently getting bitten by various workloads that query by `created_at` instead of a canonical timestamp :) Thanks, I'll take a look!

For the "valid time" primitive I was thinking of implementing something like a hybrid logical clock that CockroachDB has (but with looser guarantees, mostly just need uniqueness and monotonicity). A sequential ID would provide a slightly nicer interface for pagination but has all the problems that I previously mentioned.

Re:Waltz: A Distributed Write-Ahead Log on Monday September 09, 2019 @06:15PM (#20923070)
Here's a dumb question about log-structured systems like this: does this system work nicely with backfills? Suppose you start logging events with Waltz and you want to migrate an existing system's data into the same log. Or something goes wrong and oncall needs to manually insert old events. Does Waltz have capabilities to backfill events into the historical log or reassign transaction IDs?

This might not be needed if this is strictly used for FIFO event consumption, but I guess I was thinking of trying to make a system like this support time-sliced queries.

Re:Building a distributed time-series database on PostgreSQL on Wednesday August 21, 2019 @06:05PM (#20762373)
Right, but I'm assuming that with immutable data writes aren't a problem?
Re:Building a distributed time-series database on PostgreSQL on Wednesday August 21, 2019 @04:52PM (#20761693)
Naive question: if time-series data is presumably immutable, shouldn't it be easy to just arbitrarily replicate chunks proportionate to load?
Re:The Ruin of Britain on Sunday June 16, 2019 @04:31PM (#20197704)
I once read that King Arthur is one of the crowning (pun probably intended) counterexamples of "history is written by the victors": if the Anglo-Saxons really did write the history of the Sub-Roman Britain period, we would have heard more about the Völkerwanderung and less about Arthur and Badon. But maybe what you said is more correct, that history was adapted by the victors (who then became the losers).
Re:The Ruin of Britain on Sunday June 16, 2019 @02:29PM (#20197086)
Matter of Britain/Arthurian mythos nerd here.

It's amazing how much of the Arthurian mythologies have been derived from varius vague allusions in this book. Probably lots of people know about the theory of Ambrosius Aurelianus as a possible candidate for historical King Arthur because of his role in the battle at Mount Badon, but what's even more interesting is that in the book Gildas refers to a leader only by their sobriquet of "the bear", which was rendered as ursus in Latin, but in Brythonic, it would've been arth.

Also another fun fact, Gildas never refers to Dumnonia by name, instead calls it "Damnonia" presumably to reflect his condemnation.

Older »