Profile cover photo
Profile photo
JP Sugarbroad
I don't bite unless asked.
I don't bite unless asked.


Post has attachment

Post has shared content
For decades, American universities have been using sweatshop tactics to lower their teaching costs while layering in new high paid administrators every single year.

What a lot of students and those outside of higher education don't understand is that college teachers work 3x the hours they spend in a classroom designing classes, writing tests and assignments, grading, meeting with students, completing increasing amounts of administrative paperwork, and (if there's anything left) some of their own research.

The student response is often that adjunct faculty should just give everyone A's but you don't become a teacher because your subject doesn't matter to you. You don't go through all of this because students don't matter to you. Teachers love their professions it's just that their professions don't see them as human beings.

The response of many, if not most, full-time and tenured faculty is dismissive condescension. They've landed reputable, relatively decent paying positions so the cream really must rise to the top and all of those struggling part-time faculty are just sub par.

Further adding to this dynamic is the fact that universities cleverly pit full-time against part-time teachers in contract negotiations. What follows are rationalizations: Full-timers are often happy to believe that the status of being part-time faculty is a temporary affair because it happened to be temporary for them, or they've heard from others with that experiences.

However, most adjunct will never become full-time faculty for the simple reason that universities and colleges have been shifting class loads from full-time teachers to part-time teachers for decades. Relative to the number of part-timers employed, the number full-time faculty positions have been steadily dropping while existing full-time teachers are increasingly promoted to newly minted (and higher paying) administrative roles.

Of course, the entire edifice is built on absurdities.

At many institutions, as prestigious as you can imagine, 1/2 to 2/3 of classes are taught by part-time teachers. If part-time teachers are so inferior what does that say about the university's teaching product?

What it says is that administrators and a fair number of full-time faculty don't believe in quality higher education. It says that they believe what cynics about higher education believe, that a degree is nothing more than a culturally sanctioned ticket to a different segment of the workforce, an intrinsically meaningless piece of paper that should be produced at the lowest cost (and greatest profit) possible without concern for any other issue because profit is all that matters.

On any given day no one sacrifices more to teach college students than part-time faculty. I had colleagues who'd worked for decades as part-time teachers at several universities who were barely surviving.

There's nothing in this article that surprises me. Not one gaddamn thing. I've seen it. Your college administrators have seen it. Your full-time faculty have seen it. And most of them are fine with it. (+Joerg Fliege is an obvious exception).

via +Joerg Fliege

Post has shared content
Something for C++ developers out there: Google just open-sourced a bunch of base libraries that I highly recommend using. What these basically are is highly tuned implementations of very basic things – mutexes, string concatenation, and the like – with nearly 20 years of cleaning up all of the difficult corner cases, compiler idiosyncrasies, performance hypertuning, and so on.

Having used (and helped develop) this library at great length, I'd particularly recommend your attention to:

The entire synchronization/ folder, most especially the mutex class: this is basically the only mutex you should use in C++.

base/thread_annotations.h for some compile-time annotations that go with that, which improve your sanity tremendously (especially in combination with tools like tsan).

base/call_once.h, for running something exactly once at code startup – and correctly handling every bizarre idiosyncrasy of the language standard and compilers (of which there are more than you would guess).

All the libraries in strings, especially the ones that do string concatenation: they're considerably more efficient than string::operator+. Also, the Split() function has very nice syntax.

debugging/everything: Link the leak checker into every one of your unittests. The stack trace stuff is a tremendous help in debugging.

These classes may seem super-basic, but I can say that the amount of toil that went into resolving every obscure issue that made them complicated has been tremendous. And that toil is automatically subtracted from you: when you use these classes, things tend to Just Work.

(If you're wondering, my own contributions to the current version of the library are mostly in the strings -- most of the parts that I wrote bigger chunks of were scheduled for future versions of the library, as more stuff rolls out)

Post has attachment

Post has shared content

Post has shared content
A thorough and well-written short course in one of the parts of American history that (if your school was anything like mine) you never learned about: the Nadir, the period from roughly 1890 to 1930 where race took on its modern outlines in our country.

h/t +A.V. Flox

Post has shared content
if the Google Manifesto was correct, then you would expect to see that Google was full of mediocre female employees, who had been hired by a process biased in their favour despite being inadequate to the task. Whatever the author of the manifesto thinks, Google does not believe this to be the case and as far as I can tell from industry blogs, it isn’t – female employees in tech are generally very good. This would, of course, be consistent with the hypothesis that the current selection process is biased against them.
If, on the other hand, one had a situation where the writers of windy conservative manifestoes about not getting fair treatment were in fact mediocre whiners who inflated their CVs, then that would be evidence that there wasn’t a bias in the recruitment and retention system, and that in fact there was probably an inefficiency caused by the extent to which mediocrities were able to bump along because their face fitted in a homogeneous techbro culture. The concentration on star engineers, senior executives and Sheryl Sandberg C-Suite geniuses is entirely wrong; the progress of gender equality in the workplace ought to be measured by the extent to which women can get into the ranks of time-serving dead-wood middle management roles.

True equality will be reached when mediocrities of all kinds exist at every level. The fact that minority hires are consistently excellent is an indicator that we aren't there yet.

Post has shared content
The "manifesto": as an ex-Googler, I despise it. For the damage it's already done.

If you haven't read the Medium piece from +Yonatan Zunger linked below, I urge you to do so now. Since he recently departed Google himself just before the manifesto leak, he's (somewhat) freer to discuss this than he would have been before.

As a Xoogler, what really pisses me off is the unfair damage of this manifesto (and its leaking) will do to Google's internal culture and external image—in a way that could affect diversity efforts across the entire industry. This is something Yonatan got at very well in part 3 of his essay.

A big part of the problem with this story has been the decontextualization of the "manifesto". Let me try to raise the veil a little bit, although, due to my NDA, I must write partly in generalizations and hypotheticals:

The thing that people who haven't been or closely known Googlers won't get is the company's long history of semi-sanctioning internal dissent and minority opinions—some of which have lead to some of Google's most popular and profitable services.

At most command-and-control large corporations, the simple existence of a document like this to leak in the first place would be a terrible indictment of the company's culture, because at most companies such a document would never be allowed to even promulgate. At most companies, any potential writer would have understood—even in terms that the writer of this thing would have gotten, with his cough "male engineer mind"—posting this to be literal career suicide. Not because of its bigotry, but simply because it's semi-publicly questioning the decisions of management. ("Terminable behavior" might be the sort of thing you'd read in an employee handbook.) So there's a built-in assumption much of the media is making that for this document to even exist probably indicates it was written or sanctioned by a top executive.

But Google has always tolerated (there's that word again) smart, thoughtful "minority reports" from individual employees not speaking for anyone but themselves—many in the "<something> considered harmful" strain.

"Dissent reports" might have been more accurate terminology, because such a paper (on a different topic) might represent a view that the majority of Googlers (even in management) hold, but some other course had been decided on by the controlling managers, or Google had just fallen into a practice and it had become de facto policy without conscious thought.

I have a story about this (again, whose details I must elide). I think I first met +Yonatan Zunger when +Liz Fong-Jones and I co-chaired an officially-sanctioned group working on such a "minority" opinion, at the behest of Larry, when the executives in charge of the project in question had already declared the matter closed to discussion. (The opinion itself was first promulgated widely by an individual employee writing on his or her own behest; the officially-sanctioned committee we co-chaired was followup to that opinion.)

That minority opinion eventually largely won out, after our team worked to flesh out the particulars of the consequences of the decision, gather data, and open lines of communication. We did this partly by going out and meeting with Googlers to find out how the decision affected the company's values, and what kind of company Googlers wanted to work for (for instance, I held several "town halls" in the NYC office, where we exceeded fire-code capacity of our largest multipurpose room).

We had to try to answer these questions of company values. Was the dissenting opinion actually held by just a "minority"—specifically, of those who had informed feelings or an actual personal investment one way or another? (Decisions by default or from ignorance, chance, or due to extrinsic cultural norms were not considered valid arguments.)

If the opinion was not, in fact, that of a small "minority"—and the opinion represented a majority, or at least, demonstrable plurality of those to whom it mattered—how alienated would those who agreed with the dissent feel if the company continued down the course? Was there a stake held by the informed members of the (non-dissenting) opposing side, and if the decision were overturned, how alienated would that make them feel? These were the sort of questions we were tasked to answer.

"Minority/majority" are really the wrong words here—it was just what we got stuck with given the "minority report" terminology. In our case, it was only "minority" because there's a presumption in any business that when executives make decisions, they do so in a way that reflects company values, and company values are held by the entirety of the company, ipso facto.

Yet—to make up two hypotheticals—there's an enormous difference between a cosmetic product redesign that was well and truly hated by 90% of Googlers, and a decision about benefits affecting LGBT employees that only 40% of Googlers overall disagreed with—but that 30% had no informed opinion on, and that almost all LGBT employees disagreed with.¹ In case it's not obvious: at Google, the latter "minority" argument will win, because it's a matter of company values. The former "majority but dissenting" opinion may or may not win, because it's not about values. (Actually, it probably won't win, because Googlers are not in the business of making products for Googlers, and their opinion of aesthetic design isn't necessarily an informed one.)

At other large companies I've worked for, this entire scenario would be unthinkable—once an executive makes a decision, the only thing that will change it is that executive changing his or her mind, being replaced, or getting overridden from the top—at most companies, general employee organizing and dissent is not a valid use of company time, and will get you fired.

All that's to put into context why I hate the damage this manifesto and its leak caused. Inside Google, it will add entirely unneeded and unwanted toxicity to the work environment, especially for women, because its content is so execrable. For that reason—dissent tolerance or no—this person deserved to be terminated immediately, as Yonatan described in his final paragraphs. "Public" (inside the company) dissent was acceptable, and even, when successful or when its tough questions improved the final product, was lauded; bigotry was not.

Outside Google, because most people will be unaware of the "tolerated dissent" culture, they won't recognize this missive for what it is—the uninformed and bigoted whining of a single engineer whose support among other employees is proportionally minuscule.

I said, "proportionally minuscule." Critiques of the sexism and other "casual bigotry" of the industry by looking at the FOSS community—where things that at companies would be internal issues are done out in public—strongly suggests that there's an embarrassingly large portion of straight white cis engineers who might be sympathetic, but I very firmly believe that at Google this was less prevalent. (Not less prevalent enough to even be held up as an example—and of course its prevalence should be zero—but substantially less than other large companies I'm familiar with.)

Outside readers won't understand that this engineer had undoubtedly seen other cases of minority opinions being promulgated, thoughtfully (and sometimes passionately) discussed, and management taking action in response, and—in his own, entirely misguided way—thought he was following this notorious yet proud tradition by doing the same. That process is so bizarre, so outside the experience of employees of most large companies, that outsiders will reasonably conclude that this had to have been a semi-official document of some kind—at the very least, a message from some important dissenting executive. (And, from much of the press coverage, they seem to have assumed exactly that.)

As a direct result of this leak combined with this misunderstanding of internal Google culture, Google is going to find it even harder to recruit good talent who find these views repugnant, and will lose talent who find the increased toxicity intolerable. And that's even just limiting consideration to disproportionately overrepresented men. Diversity hiring and retention has just become a newly even worse nightmare.

More pernicious and harder to detect: it's going to attract applicants with the entirely incorrect notion that Google is a place that respects their bigotry as "just another opinion". And unfortunately, Google's hiring process—as exhaustive, selective, and byzantine as it is—doesn't screen for quietly-held bigotry. (At least, it didn't when I was privy to how it worked.)

The leak of new chief diversity exec Danielle Brown's message in response won't help matters much.² It states for the record that Google finds the views of the manifesto to be against company values, but it also restates Google's support for its culture of dissent. I think the motivation of this came from a good place, that this was meant, internally, to reassure nervous Googlers that this manifesto's leak's horrible external optics aren't going to suddenly result in a crackdown on even the good dissent, like other companies have.

Unfortunately, people without understanding of the internal Google dissent culture will read this part of Brown's message as condoning bigotry as just another argument, as appropriate as any other.

I have no easy answers here. As Yonatan wrote about his having to do cleanup—even as an ex-Googler!—this manifesto, and its decontextualized leak, has done enormous damage to the company and its diversity efforts. The manifesto-writer's being fired or not is almost irrelevant at this point except as signal—and there are important legal and ethical reasons why Google can't easily use his firing for PR purposes.

If the writer wanted a less-tolerant and less-diverse Google, he may have succeeded simply through the publicity. Undermining years of progress, as glacial as that progress may have seemed. He's managed to make Google in particular, and tech workplaces in general, more hostile unilaterally, just by clicking "share". And he probably has put a chilling effect on the dissent culture that was part of what made Google such a special place to work—which could ultimately lead to Google adopting an official policy banning internal promulgation of dissent, like other companies have.

All of that absolutely infuriates me.

You wanted answers at the end of this post? Sorry, I'm fresh out. This was despicable asymmetric behavior on the part of the writer (and possibly leaker), whose fallout at this point is still unknown and probably can only be mitigated somewhat, at best.

¹ For the record, I'm making this up—again, NDA.

² And to be clear, it won't help matters much because of its decontextualization by being leaked; I'm not criticizing her for the message's content to current Googlers. (I could wish she were more explicit that bigotry is not tolerated in the same way that other dissent is—the "paradox of tolerance" Yonatan's written about so brilliantly—but I suspect that someone else, perhaps Sundar or Larry himself, did that in a message that hasn't (yet) leaked.)

Post has shared content
So, first off, thank you to +Mickey Schulz for helping me talk myself through this one.

I disagree with this piece's conclusion. I think Zunger's piece was great, I wish more guys would stand up and say "hell no" as clearly and eloquently as he did, and (critically) I think he did so in a manner that explicitly did not steal an opportunity to speak from anyone else. I think his piece is probably getting shared more than others partly because it's great, but also partly because he's a dude, but I don't really put that on him.

However: despite that point of disagreement, this piece is a damn big truth bomb, and well worth a read. You may even find yourself agreeing with her conclusion, and if so, we're still cool. The world is complicated, and there's room for disagreement while we try to move it in the right direction.

Post has shared content
A Googler wrote an (internal, since leaked) manifesto about gender and engineering a few days ago. If you haven't read it, I will say that you are not missing much. But if you've heard about this and are wondering what my response was, I just posted it publicly.

The intro of this also hints at some bigger news which I'll get into later. :)
Wait while more posts are being loaded