HNHacker News
TopNewBestAskShowJobs

drewcrawford

4,834 karma · joined June 13, 2008

AI tooling and systems/SDK engineer · Rust, WebAssembly, Swift/Apple platforms · wasm-bindgen contributor · building AI Bot Mafia (botmafia.games) · sealedabstract.com · github.com/drewcrawford
submissionscomments
drewcrawford··on License update
Was this lawyered? Because the clause reads to me like Programmer Law.

> Provided that you are otherwise in compliance with the GPLv3... [we] also grants you the additional permission to convey through the Apple App Store non-source executable versions of the Program as incorporated into each applicable covered work as Executable Versions only under the Mozilla Public License version 2.0

1. The phrase "only under the [MPL]" definitely means my app store binary isn't distributed under the GPL

2. The phrase "Provided that you are otherwise in compliance with the GPLv3" does not seem to have any effect for distributing "only under the MPL". By distributing software "only under the MPL", I do not "otherwise" have GPLv3 obligations for me to not be "in compliance with".

3. Unless you intend to say that I have both GPLv3 and MPL obligations, in which case the same GPL obligations which [allegedly] prevented me from using this in the app store are still in effect. Also, this would contradict 1.

Please have a lawyer look at this. I guarantee my clients' legal counsel would not allow use of a library under these terms.

drewcrawford··on Freedom of speech is in retreat
As others have said, freedom of speech is not the first Amendment and America is not the world. But even American thinkers themselves define it more broadly, (and always have) e.g.

> What is the liberty of the press? Who can give it any definition which would leave not the utmost latitude for evasion? I hold it to be impracticable... [0]

> Protection against the tyranny of government isn't enough; there needs to be protection also against the tyranny of prevailing opinion and feeling [1]

> Liberty of circulating is as essential to liberty as liberty of publishing; indeed, without the circulation, the publication would be of little value. [2]

Then of course even if what you have in mind is the First Amendment, its whole historical trajectory is to expand to apply to more things. As written it applied to only Congress as distinct from any other form of government, and I note the conspicuous absence of that limitation in how you describe the freedom. And this may be news to some, but we started applying it to private actors long before I was born (Wickersham v. City of Columbia, Brentwood Acad. v. TN Sec. School Ath. Assn, etc.) So it has a long history of ever-expanding scope, and of course when you look at what e.g. Madison thought it was broad from its very birth.

Now we get to

> Well, the real world will never let you have that type of "free speech". If it bothers people... they can start another website/college that allows it. ... In other words, the non-government "free speech" has a cost and sometimes, society doesn't want to pay for it

I'm trying not to poison the well, but this same argument made here has implications for e.g. civil rights. "If you want a hotel that lets black people stay there you can start another one. Racial integration has a cost and society doesn't want to pay for it." But we decided as a society (over many folks' objections) that racial equality was important, and so we wrote legislation. Some people really disliked that at the time; some people still dislike it. However as we no longer have segregated bathrooms I think it is hard to argue that "the real world will never let you" establish human rights, it is actually the story of our country.

You are certainly free to believe that there we have no fundamental human right to expressing a political position on the Internet without living in fear of being fired from our job; that we have come quite far enough, thanks. You can also believe that one's "right" to do that is trumped by other people's rights to contract freely.

But I think in that case one should just admit that they don't see free speech as a fundamental human right. There is no shame in that position. And there is no need to hide behind an economic argument from a dark time in our history, there is no need to accuse folks who do see free speech as fundamental as "diluting" the term or that we live somewhere other than "the real world". Just saying "I don't believe it's a right" is a much faster, tighter, and more difficult argument to attack.

[0] Federalist 84

[1] JS Mill, On Liberty

[2] SCOTUS, Lovell v. City of Griffin

drewcrawford··on Traffic-weary homeowners and Waze are at war
Of course it is necessarily about tribalism. If it were not we could just install an access gate that limits the throughput of the road to X cars per hour (without regard to whether you live on it or not) and consider the problem resolved. But a homeowner would oppose having an equal share with any other citizen for a road on which they live.

Or, to see this the other way: consider a driver shrouded in a veil of ignorance, with the knowledge of what road he lives on and its traffic situation temporarily hidden from him. He sees a shortcut through a neighborhood that may be his, but it's statistically unlikely to be in his large city. He takes it.

drewcrawford··on Nix as OS X Package Manager
But how do you reproduce the dpkg packages? When your friend hands you a deb, instead of a docker image, what have you accomplished as far as knowing how they arrived at their configuration? Debs are also binary blobs, just more of them, and separately.

From what I can see, there are two philosophically pure approaches to software packaging. Either we can build everything from source, so that we have reproducibility (see e.g. emerge for a practical system, although it is still not bit-for-bit reproducible) or else we can ship everything as a blob so we have total consistency (e.g. Docker, Go).

The other approaches, apt and nix, seem to me to be compromise positions. They may even be practical or useful compromises. But they are ultimately about the degree to which we allow binary blobs, not whether we do.

But it seems to me that once you have accepted installing binaries via apt then installing them from Docker is a question of degree, not of kind. It seems very hard to me to argue that distributing binaries via Docker is really worse than via nix.

drewcrawford··on CryEngine out on GitHub
Saying that licenses should be readable to software developers is like saying sourcecode should be readable to lawyers. It's a nice idea, but in the real world, even problems as simple and universal as loading an image over HTTPS is totally impenetrable to anyone outside the software development profession.

It is the same for the law. Even the basic "email-grade" legal problems are actually very complicated, and it is no good trying to simplify them down to a lay level. I'll defer to Lawrence Rosen on this, from his book Open Source Licensing: Software Freedom and Intellectual Property Law:

> The same programmers who cringe when a lawyer attempts to write high-quality software feel no qualms about writing their own open source licenses. Their goal, it appears, is to craft something that sounds like a license, to define a form of software freedom with reasonable terms and conditions, and then wait for the community to adopt the license and distribute software under it. This technique sometimes works. Some members of the open source community are more concerned with mak- ing a philosophical statement, getting free software distributed to the world, and letting license enforcement take care of itself somehow in the future. That can be a commendable goal, but from a lawyer’s perspective, it is amateurish and risky.

The reason that BSD/MIT seem readable to you is actually that they lack a lot of clarity on important points. For example, they say nothing at all about patents, which is the primary type of software litigation today. So an Oracle-type could offer you software under an MIT license, and then turn around and sue you under patent law for using the software they offered you. Whether they would win is an interesting question, but we will not find out because you will settle.

Meanwhile, Apache2 or GPL3 or other post-1970s licenses have actual clauses to deal with this. So now your lawyer can print out the license, highlight a sentence, scribble "SEE?" below it, add "Respectfully submitted, John Smith Esq." at the bottom and that is the end of your lawsuit.

drewcrawford··on Foxconn replaces '60,000 factory workers with robots'
This feels an overly technical objection? Very little of our problems are actually "impossible" if by "impossible" you mean something like "inverting a one-way function".

What I think most people mean is that moving production is impractical, e.g., that there is no realistic incentive that justifies making the investment and so for that reason nobody will make the investment, and production will not move, etc.

It would be more effective to argue that the required investment is smaller than expected, and some particular entity does have an incentive to make the investment after all, etc.

drewcrawford··on CryEngine out on GitHub
There certainly are "respected and well-understood" "standard" texts, for example, the Apache or GPL licenses are widely understood and are used for "all kinds of applications already." The legal analyses of these texts runs into the millions of pages; we know quite a lot about them at this point.

What I believe you misunderstand is that "the aim" or "the whole concept" of an EULA is not to explain things clearly to end users. On the contrary, "the aim" of an EULA is to offer terms that are favorable to the licensor (and thus unfavorable to the end user), and this aim is best accomplished when the user does not especially comprehend the document. In other words, "the aim" of an EULA is to be obfuscatory.

Anyone who wanted to use a standard document has many options to choose from that are probably familiar to you. ISVs who have passed over the ten-or-so standard texts that you are familiar with will not be convinced by an eleventh.

drewcrawford··on CryEngine out on GitHub
I don't understand your question. "The point" of an EULA is subjective, and depends on your legal strategy, and the exact text of the document. It is generally a good idea to ask a lawyer before agreeing to anything you don't understand. I am not aware of any law requiring one to consult a lawyer before clicking next in an installer.
drewcrawford··on CryEngine out on GitHub
My Google-fu produced the following Microsoft ISVLA from 2009 (not VS-related, but this language appears in several of their EULAs from the 2000s):

http://www.microsoftvolumelicensing.com/userights/Downloader...

> Potentially Viral Software. Your license rights under Agreement and/or Academic Agreement are conditioned upon your compliance with the following terms: You must not: (a) Integrate or otherwise incorporate Potentially Viral Software into, or combine Potentially Viral Software with, any Licensed Product, Software Documentation, or derivative work thereof; (b) distribute Potentially Viral Software in conjunction with any Licensed Product, Software Documentation, or derivative work thereof; or (c) use Potentially Viral Software in the development of any derivative work of any Licensed Product or Software Documentation. “Potentially Viral Software” means software which is licensed pursuant to terms that directly or indirectly (i) create, or purport to create, obligations for Microsoft or our suppliers with respect to the Licensed Products; or (ii) grant, or purport to grant, to any third party any rights or immunities under Microsoft’s or our suppliers’ intellectual property or proprietary rights in the Licensed Products. Potentially Viral Software includes, without limitation, any software that requires as a condition of use, modification and/or distribution of such software that other software incorporated into, derived from, or distributed with such software be: (a) disclosed or distributed in source code form; (b) licensed for the purpose of making derivative works; or (c) redistributable at no charge.

I'm not a lawyer, but this term sounds scarier than I think it is. "Potentially Viral Software" is defined to be software under a license that places obligations on Microsoft (which would be every license offered by Microsoft, and no license not offered by Microsoft, so as its only effect is to prevent you from using Microsoft software, it seems an unintelligible term). It goes on to say this "includes... software that requires... [being] distributed in source code form" but that has nothing to do with "Potentially Viral Software" as defined just before.

My gut says Microsoft wrote this term in a manner calculated to confuse drive-by software engineers into believing that VS can't be used to compile GPLed software, that the GPL creates obligations on third parties, or both.

drewcrawford··on GitLab Container Registry
Does this work for autoscale runners? Are those privileged by default?
drewcrawford··on Dynamic Swift
For what it's worth, that's my preferred solution in the long-term too.

Speaking narrowly about the XCTest case:

* It doesn't actually require runtime dynamism (since you have a finite number of tests that can be statically enumerated). Following the principle of least power, don't do it if you don't have to.

* A runtime-dynamism-based solution would require `dynamic` sprinkled on function definitions which would be a departure from source compatibility with Darwin XCTest which relies on the superclass and @objc inference on methods.

* Mirror offers reads but not writes, and there is some sense that offering the ability to mutate through Mirror is uncharted design territory.

Then there is debate about whether other "dynamic" cases can also be solved with static analysis. I am in the "probably not all of them" camp, but that still leaves some of them, and it would be good to study that boundary condition in more detail.

With sufficiently robust static analysis you "can" implement your mirror trick for example (by code-generating a bigass extension to Mirror that just returns statically-analyzed data). Somebody will do that, and so I think upstream will be following how well it works out for them and letting the larger community build the case for or against the feature.

drewcrawford··on Dynamic Swift
So, I'm deeply involved in Swift, in the sense that: I write the odd compiler feature.

From my vantage point it seems to be the opposite of your analysis. While it is true that ObjC's dynamism has all kinds of problems we want to avoid, we're also running into lots of cases where the obvious solution is dynamism, and the other solutions aren't obvious.

One concrete example I can offer you is NSCoding, which is being reimplemented right now in corelibs-Foundation. That API pretty much needs a way to look up classes at runtime, (there's really no other sane way to provide the feature). You could argue that we don't need NSCoding (in fact that was argued), but ultimately, we decided we needed NSCoding more than not. So Swift 3 adds a `_typeByName` API which does dynamic lookups of classes by string. It landed, it exists, the ship has sailed.

Another concrete example that may be interesting is XCTest. The way XCTest works is your unit tests are written as functions on a class, and we "discover" the tests by enumerating the functions at runtime. But oops, Swift has no way to enumerate functions at runtime, so we can't find any of your tests. The solution to this problem is actually pretty interesting: the latest proposal is to do sourcecode analysis from the compiler to enumerate your test functions, dump that somewhere, and then generate a second program that uses the dump to call the enumerated functions.

Now that is very interesting, and certainly less dynamic than ObjC (e.g. no method_exchangeImplementations and other nonsense), and I would argue quite a lot saner for the scope of the XCTest problem. But it follows easily that as soon as we do that we could emit a program that uses that same function enumeration to emit a giant switch statement that does dispatch-by-string. Of course it would be opt-in, so only callers that wanted it would use it, but there's nobody to stop you.

All of that to say, I think what is actually happening inside Swift is we're coming up with more structured kinds of dynamism, rather than a knee-jerk reaction against the ObjC philosophy. ObjC's model certainly has its issues, and we have gotten surprisingly far without really much dynamism at all. On the other hand, Swift constantly hits cases that are "solved" by ObjC's dynamism, those are definitely real (doesn't get more real than the core libraries), and dismissing them out of hand would be stupid. Meanwhile it takes time to come up with solid "structured" approaches to all the various cases people use dynamism in the wild, so it will be a long process of enabling more and more kinds of dynamism over time.

As for the ObjC runtime, we will not actually have it for years. Maybe on Apple platforms. But the decision has been made not to ship it for Linux, and Linux is of course the next big frontier for Swift. All code that wants to run on Linux (or run on both platforms) better not use the ObjC runtime for anything ever. So this places increasing pressure on designers to figure out what our answer to all the dynamic usecases will be, because there is no seatbelt to save you on one of our platforms.

Basically, I would very much not be dismissing these critiques out of hand. They are not (other than a few) arguing for a return to ObjC. But they are raising issues not seriously solved by Swift at present, and we need to be uncovering those and putting in the design effort to generate solutions for them.

drewcrawford··on New Sublime Text update
Very few people are willing to pay the kind of money that would actually sustain a company.

Let's run the math, and see if we can build a business around a text editor.

Let's say we can get away with 2 good developers (across 3 platforms, that seems difficult to me). Let's further stipulate they are willing to take a pay cut from their previous life at Facebook and are willing to do this as a labor of love. 90k/yr salary, 180k/yr for both.

Now we need a technical writer, to handle API documentation etc. Let's say we can get a CS major to do it part-time at 20 hrs/week at $20/hr. 20k/yr.

We also need a sysadmin to wrangle the CI setup, website hosting, install updates, setup email, etc. Suppose they automated everything in Ansible/Docker/Kubernetes/whatever_the_cool_kids_do_these_days and so we only need 10 hrs/week at $50/hr. 26k/yr.

We need a QA engineer to bang on things and file proper bugs. $50k/yr. Let's further stipulate that this same person will handle customer support, because we're a lean startup and combine multiple roles in the same individual.

We need somebody to handle marketing (or maybe direct sales, since it's a $700 pricepoint). Let's call it another $50k/yr.

We're now up to $325k/yr. Traditionally we would now have to lease office space, buy macs, call our AWS sales rep, etc., but let's assume for this conversation everybody works from home, they have their own equipment, and we got free startup AWS credit, which is not very realistic at all, but whatever.

So let's stipulate this is a stable burn rate. Nobody will get poached by Facebook, nobody needs to raise a round, if we can pull in only 325k/yr, we can do this forever.

When we sell our $700/license, let's say we net 75%. That is actually very high: most software companies net around 50-60%, because the App Store, WalMart, the state, etc., take their pound of flesh. But our $700 text editor is so amazing that people will buy it direct from us, we will have a strong brand, handwave. So we take home an incredible $525.

We now need to sell 700 licenses year-over-year in order to keep this thing going. Not even 700 licenses one-time, but we actually need to close 60 licenses a month, 2 licenses a day, 365 days a year. To do that, we needa sales process.

I would love to live in a world where a salesman knocks on your door, does a 2-hour demo, and at the end you have 50% conversion to writing him a $700 check. I just don't live in that world. And that's the kind of sales process you'd need to sustainably develop a text editor.

Anything short of the exercise I just went through will result in a failed product. A developer might take a pay cut for a year but will lose interest if they're well below market. If we don't hire good QA then our quality will be shit and not worth $700, just think of all the shitty software you use already. If we don't have support then nobody will want to spend $700 to get their emails ignored. If we fund via VC then we will have to blog about Our Incredible Journey the next time Google Docs needs an engineer.

The result of this analysis is the current market. The only way to develop a text editor is either all volunteers (like Atom), a small part of a large corporation's marketing budget (VSCode), or a labor of love from one and a half underfunded and overworked people who we somehow expect to stop being underfunded and overworked even though we only paid them like half a billable hour several years ago (ST).

drewcrawford··on Save Firefox
The unsupported assertion that a DRM-free universe "is not realistic" is not an argument, even when it is repeated for emphasis in italics.

In the music industry, DRM seemed inevitable until one morning Steve Jobs woke up on the wrong side of the bed. Then the whole thing crumbled overnight.

iTunes was not "pilloried by its users for not supporting the content they want to access". Users did prefer to buy DRM-free music, bolstering iTunes marketshare in the process. The historical record demonstrates the opposite of your hypothetical.

If Chrome or Safari drew a line in the sand and said "No DRM", that would be the end of DRM on the web. It seems they are not willing to do that. But that is a reflection on the lack of leadership at the technology companies, not on the inevitability of DRM.

drewcrawford··on Lyft and Uber to Halt Operations in Austin After Voters Reject Proposal
You're very misinformed.

I live in the area. Uber/Lyft have campaigned really aggressively. I've got stuff from them in the mail, they contacted me at my unlisted cell, they've been canvassing hard for months. I've never even seen an "anti" ad.

The "pro" Uber/Lyft PAC spent north of $8m, while the "anti" PAC spent 1/40th of that [0]. This is by far the most money ever spent on any election in this city's history, by a factor of about 5. And all of it is Silicon Valley money, which is just weird.

Quite frankly I have no idea whether the regulatory reforms Uber/Lyft wanted were sensible or not. I don't know whether they were really necessary for Uber/Lyft to continue operations or not. I just know a bunch of SV investors spent an absurd amount of money trying to convince everybody of that, and Austinites are a very skeptical bunch who decided not to take their terms.

[0] http://qz.com/677908/the-future-of-uber-and-lyft-is-at-stake...

drewcrawford··on Ex-Reddit exec launches 'Imzy,' a warmer, fuzzier Reddit
I'm not asking you to generate arguments a hypothetical NSA proponent might use, from a person who assures us (s)he finds them unconvincing. Far from evidence against groupthink, that's evidence in favor of it.

I'm asking you to recount specific instances when an NSA proponent did make arguments here, that they themselves held, at the risk of their own social capital.

drewcrawford··on Ex-Reddit exec launches 'Imzy,' a warmer, fuzzier Reddit
> More importantly, on contentious issues HN will still reward quality participation.

Describe 3 good arguments in favor of the NSA you've seen on HN.

drewcrawford··on Aphantasia: How It Feels to Be Blind in Your Mind
Not really questions, but two facts I think you should know:

I read your inability to use Morgan Freeman's voice in Morgan Freeman's voice.

When you mentioned you were unable to hear the Star Wars theme my first reaction was how badly it clashed with the song in my mind's ear already.

I have a much stronger mind's ear than average. The loss of my mind's eye I would not notice much; sure it would make fiction less fun and dreams less vivid, but eh. But losing my mind's ear would be like going deaf. I have what amounts to a built-in iPod that can call up any sound I've ever heard, and many more sounds that I haven't. Going without a mind's eye would be a curiosity to me, but losing the ear would be like losing an arm.

drewcrawford··on A New Map for America
> There is virtually no way to have a consensus on what precisely needs to be guaranteed. I would personally prefer to move to a state that does not "guarantee" health insurance (I will pay for health insurance myself).

There was a consensus; we call it the Constitution. The problem is that American lawmakers today aren't able to negotiate a single law let alone a document like that.

States are not really a solution to that problem: large states have gridlock much like the feds, and small states create the patchwork of laws that prevent me from mailing you a beer but ensuring some lawyer at LargeCo will be paid to figure it out. And all states are more vulnerable to corporate capture than the feds are.

Laboratories of democracy work when we have democracies that are worth studying, and when politicians adopt fact-based policies instead of ideological ones. But we do not live in that world, and have not for at least 30 years.

drewcrawford··on “Fiercely resist any further broadening of the scope of the C UB problem”
In Swift:

    let a = Int.max + 1
Produces a compile error as the overflow is within Swift's conception of a "literal expression".

If we step out to a case where the overflow is not trivially a literal expression:

    func foo(a: Int) -> Int {
        return Int.max + a
    }

    foo(1)
This causes a runtime assertion failure in optimized or unoptimized builds.

There is a (rarely-used) "Ounchecked" optimization level for which this program produces UB.

drewcrawford··on Debian XScreenSaver package maintainer responds
> That's also an unfair characterisation

What specifically is unfair about it? Is it false?

> especially since the original report starts out recognising that exact comment.

I must be missing the part where a postmortem of this feature getting into stable was conducted and lessons were learned. Can you link me?

> Debian stable is there for a reason - not everyone is in a position where they can have the latest'n'greatest rolling distro.

Yeah, and that's why I run Debian stable. But in light of a visible lapse I'm concerned about invisible ones lurking on my systems.

drewcrawford··on Debian XScreenSaver package maintainer responds
> It's more about the spam for already-fixed features than the love of keeping users secure.

I realize this is the reason presented in the comment, but after I dug into his statements elsewhere, I developed a different picture. Anyway, there's no law that says a comment must present every argument why the code exists.

To step back a minute, I seem to be seeing the same facts very differently. Where some see an immature attempt to annoy users, I see a demonstration that Debian is unable to spot a bug when it has a 50-line comment above it complaining that Debian doesn't fix bugs.

To me, the very existence of this situation itself is a powerful argument against Debian stable as a working concept. Most of the time a bug is introduced it does not announce itself with a preamble. How the hell did this make it all the way to stable?

drewcrawford··on Debian XScreenSaver package maintainer responds
Suppose for a minute that it is legitimately impossible to have a secure screensaver older than 6 months. Like, we live in a universe where either we can have secure screensavers, or we can have Debian stable, but not both.

What we have in this situation is an engineer–as far as I can tell, an engineer on the short list of "world experts in lockscreen security"–who earnestly believes that our universe operates in exactly this way.

What is a reasonable thing to do–short of warning end users, which apparently is immature in your mind–to prevent what he believes, as a subject matter expert, to be a major and ongoing security vulnerability waiting to happen?

Should he have released under a non-DFSG license so as to prevent Debian from packaging the software at all? Or should he have politely written to the Debian maintainer asking for its removal from Debian? Should he have gotten into the Debian politics and lobbied for the "special exceptions" that iceweasel etc. enjoy to get frequent updates? Should he have taken it upon himself to backport security fixes to Debian, RHEL, etc?

I sympathize with the OS package maintainers, but I sympathize more with someone who found himself trapped between his commitment to software freedom and his commitment to keeping his users secure.

drewcrawford··on Will minimum wage hikes lead to a huge boost in automation? Only if we're lucky
> All I see is a genetic fallacy, namely appeal to authority.

Appeal to relevant authority is not a fallacy.

> because we have Austrian and Keynesian economics at battle in America.

This is a common misconception among "armchair economists". There is no battle betwen Austrian and Keynesian economics in America today any more than there is a battle between Confederates and the Union. There was a battle, years ago, but it has been over along time, although there still are a small number of weirdos still carrying on.

Now many of the conclusions of Keynesian and Austrian economics are still alive and well, but the reasoning behind them is completely different, and the war today is between camps with names like Monetarism, Lukasian Rationalist, Chicago School, and so on. The people in that poll are not citing Human Action, anymore than working programmers today are citing the 8088 specification. It is a thing that happened, it was important at the time, but we've moved on.

> The debates are not about things like the validity of the technical aspects of Adam Smith's Wealth of Nations. The debates are about whether the individual should be allowed to prosper any longer.

Have you read an economics paper? They are so technical many programmers cannot get through one. For example, here's some random abstract from the SSRN [0]:

> We provide a formula for the tax rate at the top of the Laffer curve as a function of three elasticities. Our formula applies to static models and to steady states of dynamic models. One of the elasticities that enters our formula has been estimated in the elasticity of taxable income literature. We apply standard empirical methods from this literature to data produced by reforming the tax system in a model economy. We find that these standard methods underestimate the relevant elasticity in models with endogenous human capital accumulation.

There is no debate among working economists about "whether the individual should be allowed to prosper any longer". It is like suggesting mathematicians are debating whether imaginary numbers exist.

[0] http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2688947

drewcrawford··on Will minimum wage hikes lead to a huge boost in automation? Only if we're lucky
I refer you to Question A [0], where in 2013 a few dozen practicing economists were asked this question:

> Raising the federal minimum wage to $9 per hour would make it noticeably harder for low-skilled workers to find employment.

Not a single economist chose "strongly agree" or "strongly disagree". The results were split very roughly evenly between agree, disagree, and undecided.

From this research I have derived the following conclusions:

1. Neither being for or against a raised minimum wage is an idiotic opinion from the current research. The experts are divided, so there must be no conclusive argument either way.

2. There is one real idiotic opinion: claiming to be sure. The sole point of consensus here is that nobody is confident in their answer.

[0] http://www.igmchicago.org/igm-economic-experts-panel/poll-re...

drewcrawford··on Clang vs. GCC – code coverage checking
I need to ask a stupid question. What does one do with code coverage data?

Like, I know what it is. What I'm asking is, on what occasion would the presence of code coverage data be more useful than stepping through the code in a debugger. What is something you do differently having this data than not?

Like, is there a code review gate that forbids the code coverage figure to go below 80%? Or do developers consult the reports in a self-motivated way for some reason? Regularly? What is the step between "collect code coverage reports" and "profit"?

We recently got this feature in our compiler, and I guess it seems obvious to everyone why to use it, but I must have missed that day in class. So everybody's really excited to create the report, but I can't figure out what they do with it afterward that explains the whole exercise.

drewcrawford··on Apple is bad news for the future of the Web
The citation doesn't shed any light on whether the restrictions have been effective.

It does shed light on Darwin having more overall vulnerabilities than Linux and less overall than MS, but unless you believe that browser policy specifically dominates These figures, (and can support that view) the overall vulnerability count is not relevant to browser policy.

drewcrawford··on Apple is bad news for the future of the Web
The situation with Microsoft in the 90s was substantially different. Go read all of https://www.justice.gov/atr/us-v-microsoft-courts-findings-f..., but here's a random paragraph:

> 35. Microsoft possesses a dominant, persistent, and increasing share of the world- wide market for Intel-compatible PC operating systems. Every year for the last decade, Microsoft's share of the market for Intel-compatible PC operating systems has stood above ninety percent. For the last couple of years the figure has been at least ninety-five percent, and analysts project that the share will climb even higher over the next few years.

Other than the surface similarity that they both involve browsers, the situation in that order and the present day could not be more different.

drewcrawford··on Apple is bad news for the future of the Web
What if I told you there was a valid technical reason not to allow alternate browsers on iOS?

If you're on a Mac, paste this into Terminal:

    find /Applications/ -perm +111 -type f |while read fname; do otool -L "$fname" | grep WebKit && echo "$fname"; done
With those results in mind (500 binaries for me, if you're not in the habit of blindly pasting things into terminal), ask yourself the following questions:

1. What percentage of these vendors do I trust to rapidly respond to zero-day attacks in their embedded browsing engine?

2. What percentage of these applications have I updated in the last year?

3. What percentage of these vendors do I want to grant the authority to download arbitrary untrusted internet code, compile it to machine code, and jump to it? This is what all modern JS engines do, and (other than system WebKit) absolutely forbidden on iOS.

4. Given that applications can ship "homegrown" WebKit on OSX, what percentage of programs I have installed actually do so, instead of linking /System/Library/WebKit.framework? And what does that say about the real demand for special browsing engines?

Every rant I read about how "the web is in bad shape" jumps straight into conspiracy theories about Apple the monopolist (somehow with 30% US marketshare) without solving, addressing, or even mentioning the actual technical issues.

If you're a person who does not like the way things are going, you should A) talk to people on the other team to discover what is the real, legitimate, technical concern behind our position, and B) work with us on finding solutions to that issue, and C) go out for beer afterward.

Characterizing Apple as some villain "locking up untold billions in economic value" for prioritizing security over software freedom accomplishes nothing beyond creating animosity between engineers who share many of the same goals.

drewcrawford··on I'm skeptical of NAND mirroring
"secure enclave" is a marketing term that has grown over time and is hard to speak meaningfully about.

In this particular device, the CPU has a small (~200 bytes) storage area of data that never leave the CPU, burned in at the factory, and allegedly never recorded during manufacturing. This data is involved in the cryptography, and this is what you have to brute-force. Short of capping the CPU, or some crazy side-channel attack, it's unreadable.

In addition to that, modern devices have writeable memory areas that similarly stay on-die. That (or is it the related coprocessor?) is sometimes characterized as "the secure enclave" This device does not have that. Since it lacks writeable secure storage, by manipulating the NAND you can defeat the 10-pin lockout, which for obvious reasons has to be implemented in a writeable memory. However that is different than the unlock itself.

I am pretty skeptical of the OP, as it seems to me you could just use a write-blocker to preserve the NAND, without going to the trouble of pulling apart the phone every 10 attempts. You may need to do some emulation if iOS tries to check its write, but surely our friends at a three-letter-agency already have something off-the-shelf for this.

← PreviousPage 3 of 29Next →