karljtaylor
Re:Anthropic takes legal action against OpenCode
on Friday March 20, 2026 @01:39AM (#47450897)
teams enterprise customer here, we got knocked out too. the most irritating thing here? OpenCode is necessary given how incredibly buggy Claude Code is. I know Boris is a "hot shot" but the whole individual contributor developers running rough shot on that codebase + they still haven't really fixed the cache bug makes this a really poorly timed move on their part. I can't prove this, but I would bet management doesn't actually use Claude outside of the Desktop surface. They'd have known how much this would break if they were really using it at scale.
Re:A factual warning for other media buyers
on Wednesday March 18, 2026 @03:16PM (#47430194)
this builds on earlier research, previously released (with reproducibility steps) available here: https://blog.karljtaylor.com/blog/2026-02-28-xai-search-api-...
Re:
on Wednesday March 18, 2026 @03:08PM (#47430077)
continues research I began here: https://blog.karljtaylor.com/blog/2026-02-28-xai-search-api-...
Show HN: API checks show three suppression strategies on X
Re:Slack May Have Just Published the Worst Newspaper Ad of All Time (2...
on Monday June 06, 2022 @11:29AM (#31641738)
this was a weird take then, and it's still a weird take now.
I'll admit, I'm glad we've moved past the open letter era, but as somebody that got hired to write a lot of them, I think the part the writer of this particular article got really close to understanding why this works was here at the end:
"But like too many other companies, Slack has been drinking its own Kool-Aid for too long. It seems the company believes it's the greatest thing to come along in recent years, and that it's irreplaceable. And that's a mistake. "Slack is here to stay," the ad concludes. Guess it's time to prove it."
The point of this kind of ad is, in part, to rally supporters, enthusiasts, and employees under a common goal. Given that it's 2022 and Slack is still the dominant player in the space (maybe Discord is an emergent challenge) AFAIC, it did its job.
Re:TikTok Boom
on Monday May 23, 2022 @11:35AM (#31480390)
they're going be spending significantly more than 200mn with the new revenue share program...
Re:CEO of Waymo John Krafcik is leaving
on Friday April 02, 2021 @06:13PM (#26676517)
they do, it just usually gets re-written to hell in about 30 very anxious email rounds
I've Read More Than 83 Audiobooks This Year, Here's What I've Learned
Re:University of the People: Tuition-Free, Accredited Online Degree Pr...
on Thursday June 18, 2020 @10:40AM (#23563498)
For example, one of the things we covered in my ethics class was Kant's categorical imperative. The categorical imperative falls over instantly if you apply even the most basic logic to it, but no mention of this was made in any of the course discussion or materials
I've read this a couple times, I'm curious about what you're saying here, do you mean that your class just reviewed some writing on the categorical imperative on its own, or read Groundwork of the Metaphysics of Morals?
Re:Zoom iOS app sends data to Facebook even if you don’t have a Facebo...
on Friday March 27, 2020 @01:00PM (#22704585)
yes, yes it is.
Re:Zoom iOS app sends data to Facebook even if you don’t have a Facebo...
on Friday March 27, 2020 @11:54AM (#22703965)
Vice has one of the worst privacy policies in the entirety of media, so it's kind of a curious thing to see them complaining about. They don't mention they phone Criteo and AdNexus on every page load, and I'm pretty confident I see them using Facebook events too.
Re:Launch HN: Fly.io (YC W20) – Deploy app servers close to your users
on Thursday March 19, 2020 @03:14PM (#22630641)
when companies decided it was too hard to run data driven marketing programs?
Re:Quarantine will normalize WFH and recession will denormalize full-t...
on Sunday March 08, 2020 @08:04PM (#22521655)
agreed, it goes a lot further back than ww2 https://www.npr.org/templates/story/story.php?storyId=114045...
Re:Ask HN: Which marketing plan would you choose?
on Friday January 03, 2020 @05:19PM (#21950981)
Can't quite tell where you're planning on running this, but it sounds like you're letting someone else manage the media buy for you, and you should for sure be running it yourself at this size. (The reason for this is that you want to maximize the amount you're spending on reach this early on.)
Re:Ask HN: Best visualization course in Python?
on Friday April 13, 2018 @12:52PM (#16831716)
yup! happy to.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @12:06PM (#16831302)
I could be wrong, but I've always suspected the only impact of ads like http://adage.com/article/cmo-strategy/anti-smoking-tv-ads-bi... are the padding they provide to the bottom line of the WPP shops.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @12:05PM (#16831281)
Facebook and Google (and nearly every other social ads platform that evolved along the same playbook) avoid the 'personal'by aggregating users above certain thresholds. It's fairly precise, and often the only times I can spot an individual is if they're flagging something geographic or contextual (e.g., search) that is unique. (for example, I know when my parents visit my blog because I know their network hostname and location.)
when you say, begin to search for trips to disney and start looking through photos on Facebook, it isn't that an advertiser discovers you, gnode, are looking for a disney trip, it's that the platform adds you to an audience of people interested in disney and to an audience of people who are interested in travel. advertisers engage with those audiences, platforms hold the data.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @11:04AM (#16830614)
I'm not going to argue with you, just go check Cartoon Network's media kit.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:16AM (#16830189)
targeting based on current page creates a marketplace where only those who are capable of paying an absurd CPC show up, if your stated goal is as above, it really isn't what you want.
Google had to do a lot of work on this front, and really still hasn't sorted it out very well. the personal injury keywords are a really good example of this.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:15AM (#16830177)
features are candidates, policies are selling points.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:14AM (#16830170)
it's against the rules in the US, and TV is a particularly big offender here, it's very rarely enforced.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:14AM (#16830163)
There's a handful of reasons that happens, but one of the biggest is that those ads are placed by a variety of different providers. More often than not when a large buyer goes to place a buy (think on the order of P&G promoting a new product, the snack conglomerates, movie studios etc.,) they'll place an order based on a certain number of impressions. they'll also mandate a certain viewability threshold and a number of other conditions.
In order to execute that buy, the media desk has to make purchases from a variety of platforms. Those platforms also buy and resell inventory from a mix of other places. Segments aren't very well defined, so you end up competing against yourself to buy the same inventory over again.
FWIW, a lot of those OTS services also have very traditional media desks. (e.g., https://nccmedia.com/)
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:11AM (#16830131)
except the digital, you're complaining about is also non-individual, and it's anonymized at a much higher threshold. radio points are demographics based buys, TV is demographics based...niche mags etc., same story.
Re:Ban Targeted Advertising
on Friday April 13, 2018 @10:05AM (#16830072)
I'm not sure where you stumbled into this idea, but advertising hasn't worked this way since well before the golden age of positioning...which predates the Mad Men era.
Re:Ask HN: Best visualization course in Python?
on Wednesday April 11, 2018 @10:29AM (#16811404)
Might be a little biased, but I did find this two-part series helpful. https://www.dataquest.io/course/exploratory-data-visualizati...
Re:P&G and WPP: How shit hit the fan in digital marketing
on Monday March 05, 2018 @05:01PM (#16524871)
I think it's lamar that frequently runs spots that say something like "if you see it/it works!"
Re:P&G and WPP: How shit hit the fan in digital marketing
on Monday March 05, 2018 @04:53PM (#16524815)
usually, reactions like this are indicative of being more, not less susceptible to advertising.
case in point: LiveCD is a branded capitalization.
Re:P&G and WPP: How shit hit the fan in digital marketing
on Monday March 05, 2018 @04:39PM (#16524703)
WPP != agencies
plenty of model innovations are out there, but P&G ain't that kind of brand.
Re:Ask HN: How do you overcome the problems of task switching?
on Monday March 05, 2018 @04:38PM (#16524689)
I prefer this solution as well, as I think it biases towards action, but I think it also creates a little bit of burden around communicating subtasking about projects back out at the medium level.