rlucas

20 ms·

151.
by rlucas 9y ago
Reminds me of trying to debug long-lived SSH tunnels which would fail every 2:11:15 hours. Right down to the hardcoded value in net.ipv4.tcp
152.
by rlucas 9y ago
This is a great example of why most "health insurance" (or dental or vision "insurance") isn't really insurance. It's a bundled prepayment model, similar to selling gift certificates/cards or the like. Ba
153.
by rlucas 9y ago
Daylighting, or "sunlight as disinfectant," is great in a world where shame works and the appearance of rectitude is valued. But we are seeing in US politics a world emerge where shame doesn't work, where players don't e
154.
by rlucas 9y ago
That doesn't make any sense. "Brain drain" is when your technologists leave (often going toward a more "hyper saturated VC" geography). What is left behind is the same amount of VC dollars but fewer technologists t
155.
by rlucas 9y ago
42% of the US population smoked daily. You really think that a hundred million people lit up daily with no "real" utility?
156.
by rlucas 9y ago
It's easy to dump on the management fee, especially seeing the world through the underfunded entrepreneur's eyes. But the VC fee structure is probably among the most honest and legit setups in all of finance. Look at Private Equit
157.
by rlucas 9y ago
Very good point about farm league and doing different things -- but it's not clearly the case that a VC can hold onto their Facebook or Google stock. Generally, there are restrictions in the LP agreements that may limit the amount of t
158.
by rlucas 9y ago
Clickbait title for a naive look at a pretty well-understood, much-scrutinized topic. Digging in further: - assumptions are bad. 7-8% in RE / public equities? Hmm, dubious. - poor understanding of TVM. 12% is 3x? Except, in the rea
159.
by rlucas 9y ago
There's a missing epilogue: 9) The states themselves, as well as a large interest group formed by the purchasers of whatever deals were securitized, are now financially interested in the continued financial health of the malfeasors.
160.
by rlucas 9y ago
Pitchbook doesn't even come close to figuring out the actual invested amounts and returns across an investor's portfolio. Don't get me wrong; Pitchbook is good and I pay $$$$ for it. But it serves a company-centric intel nee
161.
by rlucas 9y ago
Yes, you can buy this from Cambridge Associates, and a few other "consultants." But it costs more than you would be willing to pay out of mere curiosity. Generally speaking, the institutional limited partners (LPs; pension funds,
162.
by rlucas 10y ago
Suggested: Read Charles Murray's book on the topic. (yes, that Murray) He pencils it out to something pretty close to 1:1, maybe something like 15% nominal spend increase, but with (theoretically) much greater utility for all concerned
163.
by rlucas 10y ago
Is the assertion that adversarial ML is a subset of correctness widely considered canonical? That appears counterintuitive because so many ML techniques seem (for lack of a technically defined term) tautological. For example, a big hairy ra
164.
by rlucas 10y ago
You're confusing the fund level return with individual portfolio companies. This entire thread has been premised on the idea of getting a 10x on an individual investment. The instant matter of a 3x is a 3x at the fund level, which I&#x
165.
by rlucas 10y ago
11.6% a year for 10 years compounds out to 3x. If it was 10 years to 4x, that would be 15%. Which would be a very respectable risk premium over BH, the greatest publicly-traded large hedge fund/conglomerate of all time . But, the &q
166.
by rlucas 10y ago
This is regrettably out to lunch, my friend. A 10x return (and a nine-figure absolute return) in two years is "failure" all around? Only in the waning days of a market top would such an assertion (that such a return is a "fa
167.
by rlucas 10y ago
Similarly, once we did away with legacy electromechanical phone switches, the (phone area code) NANPA started allowing area codes with non-[0,1] middle digits. The big form of 10-digit numbers was retained and extended by relaxing a legacy
168.
by rlucas 10y ago
That probably works approximately as well as how merchants weren't allowed to store the cc number itself.
169.
by rlucas 10y ago
In finance we call the driver of this "survivorship bias." IOW, your contention is true/truthy if you ignore the fact that there is a large selective pressure weeding out the "not better" new things.
170.
by rlucas 10y ago
Discoverability is definitely a feature of any decent post-2000 *nix terminal environment. Man and apropos get you a long way. The near ubiquity of --help as a flag helps too. Well managed program distributions will even tell your shell how
171.
by rlucas 10y ago
Those numbers ring true for expectations of medium-to-large ($250 M+ is probably a good threshold) SF Bay Area VCs, at this moment in early 2017. Source: small-to-medium non-SF VC.
172.
by rlucas 10y ago
Well, except it's not straightforward, almost ever. New investment rounds dilute the common share equivalent ownership, but also add preferences and sometimes other economic rights. Also, banker fees, earnouts, etc etc. I don't di
173.
by rlucas 10y ago
It's a classic tragedy of the commons scenario. If you as a manufacturer move to a jobless production system, you gain net margin. If everybody moves to jobless production, the topline demand shrinks radically. Yet, for each individual
174.
by rlucas 10y ago
It is not uncommon in a polite social life to encounter situations where literally nothing can be done (a colleague's parent dies; an acquaintance gets cancer). Yet, doing something even symbolically is required. It is therefore quite
175.
by rlucas 10y ago
If you're sitting at the craps table then you're the only one who is likely to win...
176.
by rlucas 10y ago
Agreed, of course that even relatively mundane secrecy can serve a public interest; nuclear codes are not the only legitimate secrets. And it's very fair to say that it's unknown if the Wikileaks cost-of-secrecy effect is a net go
177.
by rlucas 10y ago
The Assange thesis isn't a binary option, secrets or no secrets. It's raising the cost of pervasive and wide ranging secrecy. A wikileaks world doesn't mean you can't keep your nuclear launch codes safe, something for wh
178.
by rlucas 10y ago
Apologies, I missed that. The review paper you linked was indeed deeply researched and had a good synthesis of what it discussed. Decided not a pop Sci listicle :)
179.
by rlucas 10y ago
Did you read the same linked article I did? There were not cites to "dozens of studies;" rather, there were quotes from the Guardian and something called Skeptoid, plus a quoted but uncited "researcher." Combined with th
180.
by rlucas 10y ago
Finally. Reviews, especially on highly contested catgories, had turned into a cesspool over the last couple years. The varieties of gamesmanship will doubtless continue but this is at least one of the hydra's heads.

More ›