Twitter | Search | |
Tweets
Dan Luu Jan 29
Both parts of this statement, that this has been addressed and that State Farm does not "engage in this sort of activity", appear to be false? Elsewhere in the HN thread, other people note that the exact same thing has happened to them.
Reply Retweet Like
Dan Luu Jan 29
Replying to @truekonrads
Do you checksum your data and scrub it regularly? Other way to phrase this is, do you think you haven't had data corruption or that you haven't observed data corruption?
Reply Retweet Like
Dan Luu Jan 25
Replying to @hillelogram
...since that changes all the time, can vary because someone negotiated a great deal, etc. Also, having a dollar value is nice because you can compare to non-hardware costs, like headcount. If that's quarter-baked, I wonder what you'd think of ideas that I consider half baked?
Reply Retweet Like
Dan Luu Jan 25
Replying to @hillelogram
I often build tools that do this! When I don't have tools that output a dollar value, I keep mind the TCO ratio between CPU, RAM, network, disk, etc., but I don't think it's reasonable to expect devs working on services to know that ratio off the top of their head...
Reply Retweet Like
Dan Luu Jan 22
Replying to @railfan_ @trav_downs
I thought it was by rakyll, but I'm not sure about that and I feel like I'm thinking of a different tweet than the one quoted. The closest one I could find was this one, but I think that's also not what I was thinking of
Reply Retweet Like
Dan Luu Jan 22
Replying to @danluu
I'd like to link to the tweet in a blog post that describes the version of the system you normally only hear about at the bar as a justification for discussing the "real" system, warts and all, but I haven't figured out the exact search I need to return the tweet :-(.
Reply Retweet Like
Dan Luu Jan 22
Can you help me find a tweet? Someone tweeted something like "talks describe an idealized version of the system, if you want to know what the system was really like, you have to talk to the speaker after the talk and public Q&A". Who tweeted that and/or what's the link?
Reply Retweet Like
Dan Luu Jan 21
This reminds me of HIPAA. Up to maybe 2010 or so, I would sometimes be told that I couldn't get a copy of my own medical records because it would be a HIPAA violation. That's literally the opposite of correct, but most medical providers basically had no idea what HIPAA was.
Reply Retweet Like
Dan Luu Jan 20
...but it's improving. OTOH, being less nasty and entitled than HN isn't a super high bar and, IMO, it's not there yet.
Reply Retweet Like
Dan Luu Jan 20
FWIW, I think lobsters comments have gotten a lot better since the moderation regime change. IMO there's still a higher fraction of mean-spritited (poster is bad and should feel bad) and entitled (poster should have written it to satisfy my preferences) comments than on HN...
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram @hyperpape
Comments don't seem to be available at because they were dynamic? That's unfortunate.
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram @skamille
That's a bad example because it would probably get some matching answers since, if you start naming cars at random, you'll name some 4d non-hatchbacks. But if the question can't be trivially answered by ignoring all of the criteria, then the asker will probably not get an answer.
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram @skamille
It usually won't be phrased with an XOR and will instead be phrased like "I'm looking for a car that has 4 doors or is a hatchback, but not both"
Reply Retweet Like
Dan Luu Jan 20
Replying to @hyperpape @hillelogram
...since the comments couldn't possibly make sense if someone had read the paper. Maybe this is an unusual example because it's a paper, but I don't think that it's *all* that different when the link is a blog post and not a paper?
Reply Retweet Like
Dan Luu Jan 20
Replying to @hyperpape @hillelogram
In general, I don't keep a record of this, but I happen to have an email written about when I tracked this. For a while, zero commenters had read anything. At 32 comments, 2 had read the abstract but it's clear that no one had read the actual paper...
Reply Retweet Like
Dan Luu Jan 20
Replying to @hyperpape @hillelogram
...it will seem as if a decent fraction of people read the post, but I think it only looks like that because people who didn't read the post get discouraged from commenting by the comment calling out the people who didn't read the post
Reply Retweet Like
Dan Luu Jan 20
Replying to @hyperpape @hillelogram
When a post is mis-titled, if you read it at the beginning, I think there's usually a pretty long time period where there are no on topic comments. Someone will (usually) eventually note that no one is reading the post, which seems to stop that, so if you look at it afterwards...
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram
But when people respond in the same way to a post by Jeff Kauffman, it's very obvious that the comment is a poor quality comment by someone who didn't read the post. Maybe this is too facile? But I think it explains why some bloggers I read get so many obviously bad comments?
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram
But for posts like PG's recent post or other widely read thought leader-y posts, comments actually seem better? What can you say to "haters are generally losers"? The posts are basically a series of rorschach tests, people can pick out any phrase and respond with their feelings.
Reply Retweet Like
Dan Luu Jan 20
Replying to @hillelogram
In the context of blogs, people who write posts with structured arguments (e.g., Ben Kuhn, Jeff Kauffman) almost exclusively get "bad" comments since a reasonable comment would require reading something more complex than a sentence with an AND or XOR, which is already impossible?
Reply Retweet Like