stevenbedrick

Just to chime in as somebody who's taught data viz for years: I didn't hear anything in your story for you to be embarrassed about! Accidentally re-inventing the box plot just means that you're thinking at the same level as Mary Eleanor Spear and John Tukey, and in my book that's something to feel pretty good about. :-)

It's a really interesting story, actually! The question of whether and when to capitalize has been an active one since at least the early 20th century. Here's a (very) recent Columbia Journalism Review article that discusses it from a stylistic perspective:

https://www.cjr.org/analysis/capital-b-black-styleguide.php

And I also thought that this article, about the Brookings Institution's decision to capitalize Black in their reports, had some interesting historical context:

https://www.brookings.edu/research/brookingscapitalizesblack...

Upvoted and agreed 100%, from an AAC researcher. Your best bet is definitely going to be to reach out to an SLP with AAC expertise.

Congress wrote the bill, yes, but in this case did so in very close coordination with the Treasury, specifically the Secretary, who presumably had a lot of input on the specifics. So it’s definitely not just on Congress, here... plenty of blame to go around.

Melanie Mitchell’s “Complexity: A Guided Tour” is a superb introduction to this area.

And while I’m plugging her books, her latest one is one of the best-written lay-audience books about machine learning and AI that I’ve seen. I strongly recommended it for anybody curious about the field, and also for people working in the area- her explanations are crystal-clear but not dumbed down in the slightest. It’s called “Artificial Intelligence: A Guide for the Thinking Person.”

In his excellent book, “The Truthful Art”, Alberto Cairo presents a great analysis and discussion of those graphs. The short version is that the way they are designed masks a lot of details of the analysis such that they are fundamentally misleading...

https://books.google.com/books?id=8dKKCwAAQBAJ&pg=PT232&lpg=...

Alternatively, please _do_ take these comments seriously, since some of the people reading this post and following this discussion are the ones who help their organizations make purchasing decisions about on-prem VCS solutions. I'll certainly be making sure that the people at my organization who are evaluating which VCS to buy are aware of Gitlab's choices regarding telemetry.

Seconding the Langlitz recommendation- their stuff is fabulous.

I loved my AlphaSmart 2000! It was a really great piece of hardware, and a real pleasure to use. I remember the feel of the keyboard very fondly...

Thanks for making such an excellent product! And I agree- there would certainly be a place for a modern version, at least in terms of functionality. No idea if there would be enough of a market, though...

It definitely happened in Portland when they ran the Orange line. We were house-hunting a year or two before the new light rail opened, and realtors showing houses along its route were already starting to make a point of mentioning it.

As soon as it actually opened, neighborhoods that could use it suddenly got a lot more desirable (read: expensive).

Agreed that it was an oversimplification, and thanks for the additional clarity and info. In the box-lifting example, doesn't that sort of thing usually have to be made very explicit in the requirements job posting?

Ah, thanks for the clarification- that's a really important point.

That sounds like a really great realization to have had!

For the record, what your old manager was doing (directly asking you about your mental health status in a quasi-interview setting) was probably illegal, at least in the US (IANAL). There are pretty strict rules around the kinds of things that employers are allowed to ask prospective employees about, and health (mental or otherwise) is very much on the "not allowed" list.

Employees have a legal right to protection from discrimination along those lines, and companies (no matter how small) most definitely do _not_ have the right to discriminate in that way.

That is an excellent analogy.

It is! And our current Librarian, Carla Hayden, is really incredible.

https://en.wikipedia.org/wiki/Librarian_of_Congress

If I may go into fanboy mode for a moment, I'd like to say that in my opinion the Library itself is one of the crown jewels of the federal government. It is also definitely worth a visit if you're ever in DC. They give free tours of the Jefferson building, and besides the beautiful architecture of the building itself, there are some very interesting exhibits on display. There's a surprising amount to see, and the staff are great- very knowledgable and friendly.

Besides its main missions of supporting congressional research and acting as the central repository library, the LoC also does a ton of great research on media preservation and archival practices... I got to tour one of their labs once- in one room I talked with an expert in the chemistry of inks used in the 18th and 19th centuries, and in the next room heard from other people who specialized in CD and audiotape manufacturing, who could talk about how specific manufacturing plants in Japan changed processes from year to year and how that affected the archival lifetime of their products. Just amazing stuff, and that wasn't even scratching the surface.

No, similar things happened at multiple agencies. The EPA was one of them.

http://www.nytimes.com/aponline/2017/01/24/us/politics/ap-us...

> WASHINGTON — The Trump administration has instituted a media blackout at the Environmental Protection Agency and barred staff from awarding any new contracts or grants, part of a broader communications clampdown within the executive branch.

It actually does do something along those lines, with the "canonical" and "compatible" equivalence rules:

https://en.wikipedia.org/wiki/Unicode_equivalence

As mentioned by others on this thread, the real issue is not with Unicode per se, but rather with the ways that web browsers handle it (or fail to handle it, as the case may be).

And that's _exactly_ why Google didn't go that route! I'm all in favor of sticking it to Comcast/Time Warner/whomever, but I very much do _not_ want my neighbors taking it upon themselves to try and run homebrew fiber through the city park by my house, or under my backyard without my knowledge or permission. If I owned utility poles, I would presumably have some opinions about people climbing up to install their own cables, and if I didn't, I'm sure that whoever I buy liability insurance from would!

Look, there are reasons why cities have permitting processes. Those reasons sometimes do include regulatory capture, etc., but generally they have to do with making it possible for large numbers of people to make productive and stable use of shared resources and infrastructure in an organized manner. I'll be the first one to agree that some permitting processes in some areas are imperfect, but that's not a reason to ignore them whole-hog.

Very well-put.

Furthermore, Github's "Education" site doesn't actually say _what_ the acadmic pricing plans are like, just that you can "ask for a discount".

A useful commentary from a linguist who studies ASL:

http://katies.online/katiesblog/index.php/2016/04/27/yes-ive...

Software like Palantir's is a classic example of a dual-use technology.

I don't know precisely, but if FERPA works the way HIPAA does, there's a carve-out for arrangements like this provided that the vendor has an agreement in place (a BAA in HIPAA-speak) and agrees to abide by certain security standards. Note that, in practice, it goes like this:

VENDOR: We promise that we are doing all of the stuff that the law says we're supposed to do to protect the covered data that you're sending us. Look, see the pretty pictures of a fancy data center in our marketing materials?

CUSTOMER: OK, that's good enough for us! (checks off box on list)

There's a third really important insight here: "Know thy product area." The author of the article may not have known anything about software, but he definitely knew a lot about call centers, as did the coders who were working with him.

In many countries, certain kinds of engineering are licensed and regulated professions. For example:

https://en.wikipedia.org/wiki/Regulation_and_licensure_in_en...

It's officially a Big Deal in some branches of engineering- structural, mechanical, and civil come to mind. My city just built a new bridge over our river, and the engineers who signed off on its design and construction were doing so under a certain legal framework that is designed to ensure that the design was done using proper methodologies, appropriate tests were conducted, etc. If the bridge fails due to shoddy design, those engineers will personally face certain professional and legal consequences. That same legal and regulatory framework gives the engineers certain protections should management try and make them cut corners on the design or testing (in theory, anyway).

People have been debating for years about whether there ought to be a similar legal framework for software engineering. There are plenty of good arguments on both sides of that discussion, IMHO, but the sorts of situation described by the original article would fall squarely under the "pro" side of the equation. I can easily envision a world in which, for software that is critical to a public safety function of government, states had to buy software written by bonded and licensed engineers. Presumably, under such a framework, a licensed software engineering company would have to have a system in place such that their "hit by a bus factor" would be >1- i.e., they'd need to have sufficient documentation or process in place to enable any single engineer to go on extended leave without preventing critical bug-fixes from taking place.

This is obviously something that any developer would recognize as a "best practice", but that in practice is often infeasible due to organizational constraints. The advantage to a system of regulation and licensure would (in theory) be that it would empower engineers to demand such practices be in place in their organizations, and would incentivize companies to set themselves up to enable such practices. In theory. As the saying goes, "in theory, there's no difference between theory and practice." Hence why this has been an ongoing debate in the software world for decades.

Yup. A long time ago, while working on a project with some particularly gnarly Unicode issues, I got in the habit of thinking in terms of grapheme clusters instead of code points (or "characters", for whatever definition of "character" one wishes to use), and it has served me very well. Combining characters pop up in the most interesting places, often where and when you least expect them! ٩(•̃̾●̮̮̃̾•̃̾)۶

Ruby's unicode_utils gem has a nice implementation of the standard grapheme cluster segmentation algorithm, and Python's wrapper around ICU works quite well. Go's concept of runes is certainly an improvement, but it doesn't handle combining characters out of the box...

Is it necessarily an "either/or" situation here? This seems to me like an example of a "both/and".

Couldn't agree more. My memories of that period are razor-sharp- you're not the only old fogie around here who remembers the aftermath of the last bubble...

When you count in the cost of benefits, retirement, overhead, etc., I can easily imagine a full-time janitorial position costing the university 70k.

It depends on the program. One thing to consider is that your funding support could be (and almost certainly will be) tied to your advisor, so changing to a better advisor will probably have financial implications.

This sort of thing is exactly the kind of thing that prospective PhD students should ask programs about before signing up, by the way. Don't be afraid to ask lots and lots of questions about things like funding, what qualifying exams are like, what happens to people who don't pass quals, whether anybody's ever changed advisors, what happens if an advisor's funding dries up, etc. etc. As somebody who's interviewed prospective PhD students, I like hearing those kinds of questions, even when I don't have good answers to them- it shows that the prospective student is doing their homework.

In "Surely You're Joking, Mr. Feynman!", there's an essay entitled "You Just Ask Them?" in which Feynman relates some truly hair-raising (in a bad way) stories about his interactions with and attitudes towards women. A quick Google search will bring up plenty of commentary on the essay, from a variety of perspectives (some of which, unfortunately, seem to view the essay as inspirational and validating, rather than as alarming).

If you haven't read "Surely You're Joking...", you absolutely should- most of the essays are not hair-raising in a bad way, and as a whole it's a wonderful collection. Furthermore, as abhorrent as I find "You Just Ask Them", I think that if one is interested in learning about Richard Feynman, one should learn about him as a complete (and very flawed) individual. The (very ugly) aspect of his personality reflected in "You Just Ask Him" was just as real as the aspect of his personality that (as reported in the original linked article) was able to connect well with a female student who had felt ignored or belittled by other physicists.

For me, that contradictory nature is part of why I find Feynman to be such an interesting historical figure.