Interesting Links for 28-09-2017

[andrewducker]

Google's self-driving car genius - and his fall from grace

(tags: Technology cars intellectual_property google )

"I Spent a Week Using Only the Alt-Right's Vision of the Internet"

(tags: internet racism politics usa )

Bulgaria passes law requiring government software to be open source

(tags: bulgaria opensource software )

An Oxford medical student stabbed her boyfriend with a bread knife. So why did she not go to prison?

(tags: law crime uk )

What happens after a defendant is found not guilty by reason of insanity?

(tags: crime mentalhealth usa )

Uber's London woes boost rival taxi apps

(tags: London taxi apps )

8 Gaslighting Techniques to Use at Work

(tags: satire work awful viaNWhyte )

What You Call a Color Depends on How You Use It

(tags: colour language psychology )

Saudi school minister fired after textbook shows Yoda at UN signing ceremony

(tags: saudiarabia StarWars photoshop history fail )

The trailer for "Annihilation" looks really good

(tags: movies trailer video scifi )

Oxytocin, Vasopressin, and aggression in dogs

(tags: dogs hormones aggression behaviour )

Why you shouldn't worry about your personal IQ score

(tags: iq Intelligence psychology success )

Earth's creation was a 'cowboy-building job' that means this planet is unique

(tags: earth space )

Germany is successful and happy because its values are the opposite of Silicon Valley's

(tags: society germany politics economics happiness )

Why do we put up with software being so badly written?

(tags: regulation software engineering epicfail security )

Comments

Why do we put up with software being so badly written?

[jack]

For the first 40% I honestly thought this was about bridges. I really hope that *now* we're quite good at building bridges that don't fall down. But I think there's plenty of physical structures that are built really shoddily, until we've accepted the necessity of "a lot of bureaucracy".

Re: Why do we put up with software being so badly written?

[andrewducker]

Yeah, I think that at least _some_ software is now at the point where regulation is mandatory.

More to come though, I suspect.

[calimac]

Self-driving cars: Wow, yet another long feature article about a totally sinister and amoral high-tech creep.

Stabbed her boyfriend: The tenor of this article is that she didn't just get off because she was well-off and white, that there were good reasons for it. But that proves nothing by itself: some lawyer could make an equally good argument that some other otherwise identical perp who happened to be poor and black deserved to have the book thrown at her. The real question is equity between cases, and if the writers don't address that, they're missing the point.

Taxi apps: I saw some other article on the Uber ban that listed about 6 or 8 London taxi apps that I'd never heard of. One of them didn't even require a smartphone: you could also book using a website.

Gaslighting: Another one that was actually used on me was, "Invent unwritten norms out of thin air and tsk at your colleague for being unaware of them."

Software: I too really thought this was about bridges. But it's true of bridges. I mean, given enough time, an untended bridge will eventually fall down. The author intends it as an outrageous metaphor, but it doesn't really work.

[jack]

"Invent unwritten norms out of thin air and tsk at your colleague for being unaware of them"

Yeah :(

[danieldwilliam]

The author of the piece on Lavinia Woodward specifically addresses equity between cases. Their comment on that is
a) this particular case has not been treated outside the norms of the sentencing structure we're just being told that it is to sell newspapers
b) you don't hear about cases where "ordinary" people or BAME culprits are treated the same as Ms Woodward because they don't have the front-page outrage capacity of Toff Justice! One Law For the Rich!
c) The author isn't denying that BAME culprits are likely to be treated less favourably. They only assert that Ms Woodward hasn't been treated outside the sentencing norms that judges are also supposed to apply to BAME culprits.
d) They specfically mention the Bashir case, where a similar application of the guidelines in a case involving a BAME cricketer resulted in a similar reduction to the sentence - until it turned out that Bashir had lied to the judge about his cricketing prospects and got pulled in for re-sentencing and a stern telling off

And I don't think they are missing the point - the point they are making in this article and consistently in other articles they have written on sentencing isn't that everyone actually gets treated the same or that the sentencing guidelines are correct but that in cases that are sensentationalised by the press as too lenient often the judges have correctly and consistently applied the existing sentencing guidelines and arrived at what the law says is an appropriate sentence.

They aren't addressing equity between cases. They are interested in conformance with the actual law.

Ms Woodward hasn't been treated more favourably than the law says she ought to be treated. A BAME culprit who had similar prospects of a job and crucially had shown genuine remorse and genuine determination to address the root cause of her offending should expect similar clemency.

[calimac]

Seems to me that what you're describing exactly defines missing the point.

Noble declaration that this case conformed with the law is completely disingenuous if other cases, especially involving BAME defendants, did not conform with (this interpretation of) the law.

The author is trying to ward off objections that the Oxford defendant got off unfairly. What fuels the protests which the author is trying to defuse? Answer: perceptions of inequity between cases: what the author would call not "conforming with the law." Does that perception reflect any reality? What is the pattern of sentencing here, in a broader selection of cases? That would address the point.

[danieldwilliam]

I think I understand what you think the point the author *ought* to be addressing is but I don't think they are aiming for the same mark.

And it's your remark about what fuels the perception which is relevant.

Having followed the series of articles by the Secret Barrister on sentencing for some time what I think they are aiming at is this; the press, in order to sell papers, and politicians, in order to garner votes, stir up outrage by suggesting that people who are "Not Us" are being treated *outside* the scope of the law when sentenced. By examining a couple of headline grabbing cases they are attempting to demonstrate that in fact all of these cases are being treated well inside the law. There don't appear to be any gross misapplications of the law.

They are not attempting to prove that there is no prejuduce in the English criminal justice system. They are attempting to demonstrate that the press and politicians can not be trusted to comment correctly on criminal justice matters and are manipulating us for their own financial and poltical aims. Is wide spread racial prejudice in the operation of the English criminal justice system a bigger or smaller story than the wholesale manipulation of the general public by headline writers for political ends? I guess that is a matter of taste but I think it is clear which the Secret Barrister is aiming at. Prejudice and propaganda are different things.

I'd track back through to the Bashir Case if you are interested in the handling of the treatment of BAME culprits.

[calimac]

Sure, the author is aiming at something different than I think they should be aiming at.

And what I'm saying is, that what they're aiming at is missing the point.

I'm not going to track anything. I need someone who's knowledgeable about the law to do the tracking. This author is, but is failing to do it.

The Bashir case cannot help me. Its facts are too different. I can't tell if there's equity between white and BAME culprits if there are too many other different variables; I need a multiple-case survey of otherwise roughly identical cases.

[andrewducker]

Yeah, some places are already good for taxi apps. The Edinburgh taxi company I use lets me book by phone, app, or website, and works fine for my needs.

[agoodwinsmith]

Just reading about gaslighting makes me completely furious. This is partly because I am not verbally nimble enough to respond in kind. I've worked with people who can respond in kind and carry on with their original statement, shutting out the gaslighter. It is very satisfying to witness.

[andrewducker]

It is.

I'm hoping that this kind of thing makes the rounds more, and that eventually people stop being able to pull this shit.

[heron61]

Earth's creation was a 'cowboy-building job' that means this planet is unique

I'm not that convinced - I could easily see this reflecting the fact that as a planet grows during formation, its composition changes, and because Earth has a mass equal to more than 9 times that of Mars, the composition is quite different. Also, Mars is (we think) in the outer edge of the Sun's habitable zone, while Earth is (we think) near the inner edge, which I would presume would make notable differences in composition. It's not like we have anything remotely like a useful data set for any of this.

bridges

[darkoshi]

A part of me wants to argue that software is not at all like bridges. And that most of it works fairly well for most use-cases, considering how many different things it has to do. Even when it doesn't work well, it generally doesn't have fatal consequences. Software that can have life-and-death results does have to go through much more rigorous testing and verification (at least, I hope it does). Criminals don't generally go around trying to make bridges fall down or catch on fire, or to intercept and hijack the vehicles traveling on them. During wars, when the military wants to make them fall down, it isn't usually that difficult - drop a bomb. So it's not like bridges are hack-proof either.

But in spite of all that, I do agree that a lot of software could use a lot of improvement. But are consumers willing to pay for the extra quality? Especially when the user interface and functionality would be mostly the same either way, with only the behind-the-scenes parts different? And considering that most software that people use, they don't directly pay for anyway? I just don't see it happening. Although there is a good argument that cleaning up spaghetti code makes software easier to maintain and cheaper in the long run. And that some security requirements might be mandated, if there was a standard way of testing software for those requirements. But the requirements and standards would be constantly changing, so it wouldn't be simple to implement such mandates.