jsomers

Yeah, and it's really worth checking out https://dynamicland.org/, because Bret Victor is actually doing this -- slash pointing the way to what such a world could look like. It just seems like now might be a good time for specific smaller parts of that vision to be carved off and developed further. I say that largely because of the advances in multimodal AI, which maybe haven't been fully applied yet in this area.

This was posted when it came out here: https://news.ycombinator.com/item?id=45802029. It generated a lot of comments -- more heat than light, possibly -- and I wonder if instead of just taking the title as a jumping-off point, folks could engage with the meat of the article itself.

(I wrote the article. I'm a longtime HN user. I find that threads here lately have gotten very jumpy-offy -- commenters use a specific article about e.g. icebergs melting to have a conversation about climate change and climate change denial, instead of to talk about the merits of the particular article -- and I was hoping to nudge folks to read the full piece, then comment on specific parts of it. I'm not sure that'll work but figured it's worth a try!)

Hi, author here. This article was many years in the making. It's ostensibly a story about reading minds but really it's about the unreasonable power of high-dimensional vector spaces.

That made it pretty tough to write: how do you explain dimensionality reduction, PCA, word2vec, etc., and the wonders of high-dimensional "embeddings" (of the sort you find in deep neural nets) when a lot—or all—of these ideas might be new to the reader? I'm not sure—but this was my attempt!

Good question! I would feel free to use the doc itself as the forum—maybe just create a section at the bottom for general comments.

I didn’t mean to say that nobody uses the software, just that I don’t know anybody who does. Which just makes it hard for me to get a handle on its value.

The question I’d ask those physicists is whether Mathematica does the kinds of things Wolfram claims for it / does things other programming languages can’t do. I wouldn’t be surprised if the answer was “yes,” I’m just not sure.

It’s hard to think about Wolfram. His tone is so off-putting — he’s constantly discovering the glorious and singular capabilities of his own products — but for instance he did basically invent the now-ubiquitous idea of the computational notebook.

Sadly I don’t know a single person who has used Wolfram’s software extensively. I haven’t used it myself extensively. Is that because it’s not what it’s cracked up to be, or because it’s a closed world?

I recently built a little trivia game for my friend, and not wanting to spin up a database, I used Google Sheets as the backend.

My friend can edit the sheet at will, and the app pulls the data in via JavaScript. Google already sets the right CORS headers.

I started it somewhat arbitrarily in 1900.

Occasionally you see articles on HN with a date in the title, like “(1998)” — and over the years I’ve noticed that these tend to be some of the best posts.

It makes sense: on a site devoted to news, an article posted so long after it was published has to be especially good.

So I hacked together this page, which links to every HN post with a date in its title earning more than 40 votes. It’s sorted in chronological order to encourage wandering.

That's a cool device, and I thought about getting one -- but for me, I wanted the satisfaction of putting ink to paper.

"Whoever wrote this needs somebody to take the fall. And that's Phreak, and that's Joey, and that's us." –Hackers (1995)

The way I think about it is less as an encyclopedia of sequences per se than it is a kind of projection of mathematics onto sequence-space. The point being that an integer sequence, because it often has so many mathematical interpretations, acts in some sense like the intersection of those interpretations. There are of course many ways in which two or more mathematical concepts are linked, but a shared integer sequence is among the most useful, precisely because it can be browsed and searched like an encyclopedia.

you're on your honor to actually guess :), i.e., to type enough of the name that there won't be a false positive

It's crowdsourced by our users, and vetted/edited by our moderators (also part of the community).

Nice!

Curious to know what you're using to parse the essays into sentences. Isn't that quite difficult in the general case?

Here's a little Ruby implementation I wrote a while back: https://github.com/jsomers/solitaire. I hope it's correct...

Jon Gertner's recent book about Bell Labs talks about this. Having a monopoly was essential for the Labs' success. A few choice passages:

Still, the contrasts between these organizations and Bell Labs are crucial. “This was a company that literally dumped technology on our country,” the physics historian Michael Riordan has said of Bell Labs. “I don’t think we’ll see an organization with that kind of record ever again.” The expectation that, say, Google or Apple could behave like Bell Labs—that such companies could invest heavily in basic or applied research and then sprinkle the results freely around California—seems misplaced, if not naive. Such companies don’t exist as part of a highly regulated national public trust. They exist as part of our international capital markets. They are superb at producing a specific and limited range of technology products.

...

“I’ve often said to my old friends that we were very lucky we got to work there, in an environment that I don’t think will ever exist again,” remarks Dick Frenkiel, who worked on the first generation of cellular technology. “It’s hard to say something will never happen again. But with the monopoly gone, with the whole concept of monopoly essentially discredited, how could there ever be a place like that again?”

...

That perceived natural monopoly wasn’t only justified by the phone system’s technological complexity and interdependence; it was also—in an argument that telephone executives made over and over again—a matter of economics. With one company in effect serving the country’s phone customers, some parts of the phone business that were highly profitable, such as long-distance service, could subsidize other aspects that were less profitable, such as local calling. Profits from high-paying corporate customers, moreover, could subsidize service to residential customers. Profits from dense urban areas could subsidize expansion into sparse rural areas. All in all, this kind of “averaging,” as it was sometimes called, helped make telephone service available and affordable for most Americans.

...

In testimony before the U.S. Senate’s Subcommittee on Antitrust and Monopoly, Bill Baker asserted that “the notion that Bell Laboratories could endure and function away from AT&T, Western Electric, and the operating integrated Bell System would be laughable were it not so sinister and so ominous.” It was an argument like the one a gifted child might make in favor of preserving his parents’ marriage.

I wrote a response to this (in defense of philosophy) a few years ago at http://jsomers.net/pg-philosophy.html.

And here's the yc discussion about it: http://news.ycombinator.com/item?id=404707.

I think jlees makes an excellent point, that PE manages to bridge the gap between the basic "hello world" stuff and the more "algorithmic-y" aspects of programming. In my case that was what really got me snowballing toward being a reasonably capable coder.

I do think you have to be excited about discrete math to really plow through the problems, but the nice thing is, discrete math is the easiest kind of math to get hooked on, just because it's not much more than fancy counting.

I am skeptical of advice from people who disagree with what I stand for.

As a rule this is a pretty impoverished epistemic strategy.

To a certain extent, yes. After the "it turns out" post I think some people went around to other parts of my site and started submitting different posts to HN. There was one point where I had three posts simultaneously on the front page (it was a glorious moment).

That converted into about a doubling of my RSS subscriber base to just over 100.

But culling RSS subscribers seems to be mostly a game of posting and linking and getting buzz on close a daily basis, whereas I'll occasionally have one or two months of inactivity.

Otherwise, somewhere between 0.5 and 1% of visitors ever check out other posts.

Google Docs will show you a diff, but without your comments. They also don't have a live preview of what your changes will look like. It also takes a lot more clicks (and time) to generate the revision history.

The overall point being that Google Docs does a lot more than I need. YMMV.

I seem to do a lot of copyediting these days (of friends' papers, application essays, stories, letters, and the like), and keep wishing there was a super-simple way to track my changes and present them to the writer in a visually intuitive way.

This "diffly" app is my first cut at a solution.

It uses Google's diff-match-patch to do the diff. There's no toolbar or lasso or any of that - comments use the Markdown hyperlink syntax, and insertions/deletions are tracked automatically.

I thought I was seeing all these Egypt Twitter stories because I get most of my news through nerd-channels like news.yc, but I s'pose they've invaded mainstream coverage as well.

My initial reaction is to agree with Gladwell, even to go further: I want to say that what we're doing when we focus on the "how" instead of the "what" in stories of modern mass communication is fetishizing indisputably "authentic" uses of technologies whose trivializing influence we fear or are at least insecure about.

But new platforms like Twitter could plausibly be changing the dynamics of political revolutions (etc.) in such significant ways that it really is worth talking about above and beyond the ground-level facts of the stories themselves.

I think Gladwell's just exhausted by the conversation.

There is a wonderful book edited by Carl Sagan called Communication With Extraterrestrial Intelligence (CETI), Proceedings of a conference held at the Byurakan Astrophysical Observatory, Yerevan, USSR, 5-11 September 1971.

The chief aims of the conference were, first, to estimate values for terms of the Drake equation, and second, to devise strategies for contact. So all these top American and Soviet scientists had to discuss a remarkable range of topics—everything from the chemical origins of life, to the anthropology of early civilizations, to the problem of encoding messages in a form comprehensible to intelligences vastly unlike ourselves.

It's possibly the most interesting thing I've ever read.

I do have a local instance of etherpad but the friction to get it running seems to be just enough to prevent me from firing it up. So mostly I use either textmate or something called writeroom. I like using textmate with markdown because you can highlight something, hit ⌘I, and it's italicized.

For a while I thought where I wrote mattered, but after producing good and bad just about everywhere I decided to just go with whatever jumps to mind first.

I'm amused when I look back at my "breadcrumbs," but I don't know if they're useful. They're just a record of a struggle you already went through. So it's not like you're going to learn a whole lot of new stuff.

Where they're useful is when you think to yourself, "maybe the way I phrased it was better." I did that a number of times, and probably wouldn't have been able to if I hadn't used etherpad.

It's really tough to say. I bet a lot of the benefits would accrue over the long run.

Good catch. What mostly happened was that an old prof of mine, upon reading a draft, pointed me to a much better source than the ones I was using. So I ditched that material (and those citations) and just used the Rainey article, the one footnoted at the end of the final draft.

That's a hell of an idea. One thing it took me a long time to learn was that being good at math is mostly about working really hard. It's hard to know that, because until you're a graduate student the only messy incomplete work you ever see is your own. Everything you find in textbooks is pristine, and worse, it all looks like it occurred quite naturally to the author.

I'm curious: how do you know how long it takes writers of successful blog posts on HN to write those posts?

Your number feels right—I've written a number of "successful" HN posts myself, and they've all taken about 6-10 hours to write—but I wouldn't want to make the kind of statement you've made, that "for the most part this rule seems to hold well," until I've seen lots and lots of data.

Exactly. I like what the mathematician Paul Halmos had to say on the same subject:

Archimedes taught us that a small quantity added to itself often enough becomes a large quantity (or, in proverbial terms, every little bit helps). When it comes to accomplishing the bulk of the world’s work, and, in particular, when it comes to writing a book, I believe that the converse of Archimedes’ teaching is also true: the only way to write a large book is to keep writing a small bit of it, steadily every day, with no exception, with no holiday. A good technique, to help the steadiness of your rate of production, is to stop each day by priming the pump for the next day. What will you begin with tomorrow? What is the content of the next section to be; what is its title?