jmrobertson

7 ปีที่แล้วdiscuss

yup it's all about that empathy, and having a constructive, vs. destructive approach to candid criticism and conversation. SO somehow swung far in the destructive direction. Candid discussion does not have to be toxic to be direct, no matter what Linus and co might say, and SO seems to embody.

jmrobertson

7 ปีที่แล้วdiscuss

I have it just for the small form factor. Luckily bought it 2 months before it got killed. Good news is though based on a recent screen crack replacement, Apple still supports fixing it.

jmrobertson

7 ปีที่แล้วdiscuss

You can do your own Big Short by buying deep OTM puts. Easy enough to research/google from there, but it can be CDS for the layman.

jmrobertson

7 ปีที่แล้วdiscuss

on-boarding new devs from the bootcamps, transitioning them to AWS + Lambda is as if starting from square one in terms of what they were expecting to work on. Very much a challenge, especially related to getting them to think about how to leverage good CI/CD for the cloud. Still a lot of knowledge gaps out there

jmrobertson

7 ปีที่แล้วdiscuss

yeah I wonder

jmrobertson

7 ปีที่แล้วdiscuss

>the harm isn't consumer oriented, since it's somewhat diffuse, it's about concentration of market power in the industry.

That's true, but that's not how anti-trust law works in the US unfortunately. It's consumer-harm based.

jmrobertson

7 ปีที่แล้วdiscuss

I don't disagree with you so not sure your point, see my original point. I'm stating that the Tesla comparison doesn't apply to what Apple's specific situation is, and therefore what policy proposals/considerations apply.

jmrobertson

7 ปีที่แล้วdiscuss

I obviously have the view that the walled garden benefits consumers, so my bias shows. And, I think there's a lot of wiggle room on the cost to host in the App store.

However, US monopoly law is based on consumer harm, so...

I think on a mobile platform, with plenty of competition to not buy an iOS-based phone, a walled garden absolutely benefits consumers more than it hurts. There is so much PII on phones now. Given the total lack of InfoSec knowledge, especially at the mobile-user level, a walled garden is crucial: see every Google Store vuln that hasn't hit Apple.

jmrobertson

7 ปีที่แล้วdiscuss

Well, Intel and AMD have had the run of the marketplace (and ARM indirectly) since the beginning, but there's reasons for that both security and other so not exactly a good comparison.

jmrobertson

7 ปีที่แล้วdiscuss

To be fair, that's inaccurately reframing the argument. Apple with an Apple store for Apple Apps is the equivalent to what you're saying. Tesla doesn't sell Teslas _and_ BMWs.

jmrobertson

7 ปีที่แล้วdiscuss

that gets into interesting Human-Comp Interaction discussions, but a good place to start is that the HCI dynamic of mobile/tablet platforms is totally different (deliberately) from that traditional approach to browsers. So I guess it could be done, but relying on that approach undermines a lot of what make mobile platforms 'mobile platforms.'

jmrobertson

7 ปีที่แล้วdiscuss

Right, so imagine separating Apple/Google totally from its current oversight position, and having a totally open app store. Like the world _just_ got past learning not to click 'download here!' on a browser, and that was after 20 years of the internet. I'd be open to alternative proposals, but separating the App Store with no replacement is not a secure solution by any means and causes more problems than it solves, assuming a secure app store is the most important trait of an app store.

jmrobertson

7 ปีที่แล้วdiscuss

Reading her stuff, it seems like proposals to separate Apple from the App store indicate a total lack of awareness about InfoSec, and at best a poorly researched policy proposal from a technical standpoint. The only reason the App Store isn't a simmering cesspool of malware is that Apple heavily moderates what gets on there, and Google does somewhat the same. That doesn't mean that there perhaps isn't a less MSFT + IE type of solution, and I don't necessarily disagree with her, but her not mentioning that sort of nuance at all makes it rather clear the policy doesn't extend too far into anything more than populism.

jmrobertson

7 ปีที่แล้วdiscuss

That's a great point. I agree that this will sort of resolve itself, until MIT puts out some new Cat 8 ethernet or some major hardware jumps, the returns start plateauing quickly. When there's such a large barrier to entry, and with only a handful of HFT/DMMs moving the major volumes, that's somewhat of an systemic risk to the tune of a SysAdmin/DevOps guy screwing up a deployment, i.e. what took down Knight Capital

jmrobertson

7 ปีที่แล้วdiscuss

Yes, not shady, but it totally unravels the argument concerning HFTs helping price discovery. I know the "hey retirement and index funds are doing it, think of the pensioners!" argument is common defense of it, but that's not at all the target audience of DPs. It's liquidity going off exchange, which hurts price discovery, simple as that. Considering how competent trade execution/slippage capabilities are for the types of broker-dealers that would be handling index/ret/pension volumes, that argument is nonsense

jmrobertson

7 ปีที่แล้วdiscuss

Ehhhhh. yes. But that glosses over dark pools, some of the shadier exchange infrastructure arbs that go on, etc.

To be candid, I'm fully on the same page as you: what's going on here is no different than what a really good pit trader did from 1800's -> 2000. It's just the digital, semi-invisible version, and because volumes are so large, the money is exponentially larger as well.

But, that's not all the HFTs get up to; a lot of it doesn't hit public markets, or isn't particularly helping w/ efficiency.

jmrobertson

7 ปีที่แล้วdiscuss

The issues with any algo trading strategy, is if it becomes known, then the trade gets crowded out and priced away. The logic really holds that if you know a shortcut to work that saves 60 minutes on a 2 hr commute, and then everyone else found about it and did it too, goodbye secret commute. This is why the strategy aspect of this gets kept very quiet by R-tech, JS, etc.

jmrobertson

7 ปีที่แล้วdiscuss

Yeah it's always worth remembering how heavily fundamental engineering work processes come into play. Knight Capital went down for a manual deployment error, of all things

jmrobertson

7 ปีที่แล้วdiscuss

https://meanderful.blogspot.com/2018/01/the-accidental-hft-f...

My all time favorite article related to this topic, I always try to promote it Guys like the author started this whole rodeo through some really ingenious hardware/software/networking jerryrigging.

jmrobertson

7 ปีที่แล้วdiscuss

Well yes, those instances would be companies that aren't on the train quite yet of understanding what a toxic worker actually is.

Which is the whole point of these sorts of studies: education

In areas where "toxic working" is ID'd as the risk it really is (i.e. this study is old, old news), it's never a term thrown around as a political term like you're concerned about. It's a specifically cancerous, horrible way of working that's seriously easy to identify. It's like trying to play politics with the term "water leaking" to describe anything other than water leaking.

comments