Say something about it.

It is this "denial of justice" provision that is perhaps most worrisome. There are many examples (here and here) of how this seemingly innocuous provision, buried at the end of this otherwise reasonable list of protections can be a real problem for citizens who will bear the cost of these lawsuits.

Even the European Parliament voted in its resolution of 8 July 2015 to "replace the ISDS system with a new system for resolving disputes between investors and states"

Test

What a great guy.

Actually, there's some great information about this.

Like this.

We should look no further than the European Parliament when wondering whether the ISDS is an effective system for dispute resolution!

Background here

This is a key example of how the principle of "Denial of Justice" in TTIP can be quite problematic. [Source article]

Good background document on the Yukos v Russia ISDS case.

This is a great example of the problems with TTIP.

Translated:

"Vattenfall had sued Germany in 2012 after the atomic got out damages amounting to nearly 4.7 billion euros to an international arbitration court in Washington. In addition, Vattenfall can make default interest per annum over four per cent, or around EUR 190 million claim, if successful, but still. "The federal government currently expects a hearing in the summer of 2016", according to the response."

Test

Yeah, well, she was also top-roped. This was a climb any kid could do with zero risk. Humorous, but not particularly climb-worthy. And-- was this a stunt any cordless vacuum could have handled?

Still -- got my click.

About 25% of the time I am approved for PreCheck despite never having enrolled. Is this because I've been a United MileagePlus member (Gold presently) for close to 20 years?

If I'm clearly a low enough threat (native born, long time frequent flyer, no criminal record) that they occasionally give me the go ahead without having done any of the additional screening questions, then perhaps they should just use these same metrics to auto-enroll a large part of the US population.

Why as tax payers are we being asked to pay for "access" to something that could be free to begin with? Particularly if they're trying to reduce long lines and deal with staffing levels-- not to mention saving travelers the time and hassle of the longer process?

I'd probably vote for "incredibly naive" over "a little bit idealistic". Clearly the assumption is not that the average reader will go to the CF website to read up on the alternate take.

Rather, what we want to achieve is to slowly begin to put more and more pressure upstream on the journalists themselves and those who publish their stories.

This will take more coverage of stories, more quickly, and over many more domains than just climate science.

test

This.

Like this.

Hi Paul.

Indeed we do!

For the time being. Our next steps are to bring annotation to publishers, so that the annotated layer can be seen by default. We've made progress on that through the Annotating All Knowledge coalition that we announced w/ 60 scholarly publishers in December, which Andy mentions further up.

Long term the solution is to bring annotation into the browser, so that it's there natively for all of us by default. This is the work of the W3C Web Annotation working group but it will take some time. There's much to be done.

Testing!!

This is a test.

A close parallel is the situation with movies and music. The online services like Netflix and Spotify are getting good enough now, that people often use them now instead of pirating via Bittorrent. Open access is the scholarly equivalent.

So true. Getting the annotation of pages right is incredibly challenging-- we do have a complete overhaul of the stream planned though. There are some great things coming!

Today Marc Perton at Newsweek covered Hypothes.is with a great article. http://www.newsweek.com/web-annotation-can-comment-any-page-should-you-452666

There seems to be some speculation that perhaps he had a long term drug addiction to opiates that was complicated by recent prescriptions to treat a hip disorder. Sheila E dismisses this though...

What is the Web Annotation working groups thinking on this? Is there anything that the WG needs to formalize?

So, clearly 221 is 13x17. But I wonder why an overlapping year doesn't happen more often. I guess neither 13 nor 17 year cicadas have alternate populations that pop up in off-years? Why not? Is there an obvious evolutionary reason?

It's not as simple as this.

First, what I expect folks’ like Dawson are saying is "don’t make public annotations here”. Are they really arguing that I shouldn’t be able to make private ones either? Probably not.

But what prevents folks from abusing such a button by disabling annotation on things that are of public importance? What if the corrupt government of Turkey puts a “don’t annotate this” tag on their entire gov’t website?

Hypothetically, how would annotation service providers distinguish between respecting such a tag when perhaps it’s a blog by a woman such as Dawson, vs a government like Turkey. It might seem simple conceptually, but there are billions of pages out there, and limited staff. What about a blog by a [insert your favorite scientific or social issue] skeptic who doesn’t want expert community to fact check their blog. Should annotation providers respect that wish and deny this powerful critical lens to concerned scientific or humanist communities?

A slight misquote: What I said (or meant anyway) was more that when humans reason alone, they're more likely to engage in things like self-dealing. A good primer on this is John Brockman's piece in the edge where he explore's Mercier and Sperber's theories on this.

... for now? We've planned extensions for other browsers.

Simplest perhaps... a mobile browser would potentially be another option for heavy annotators.

Well, to be fair, statements by either the NYSE or the federal gov't about whether the halt was a hack or not probably aren't worth much. Public statements like this are pretty much always a calculation of desired messaging and likely reaction, not some objective reflection of truth.

I completely disagree. OKCupid did the right thing.

Say something.

Test

Wait... small chihuahuas!? That can't be right!

The terror of his situation is causing him to feel weak and vulnerable. It’s strange that he should say it’s draining his lifeblood, because fear typically increases heart rate to redistribute blood to the muscles, thus priming the body to fight. However, when fight-or-flight has an alternative: fear paralysis. In infants, situations of extreme terror has triggers the fear paralysis reflex and may induce SID.

Discussed by Jain (2016) here.

This is a key point that relates to another by Dutta and Sorbo (2016) on pNGB quintessence.

Test

This clearly relates to this other blah blah ....asdfapsdjf;lasdkjf;lasdjkf'asd;fasdlkjf;askdjf;asjdkf;kasdjfl;kajsdfasdkf

So true. Further, the culture of the teams behind these platforms has more influence and power in determining the cultural norms of the world we're headed to than nearly anything else.

This might be one of the more brilliant team names ever.

Yes, annotation tech, either Genius's, Hypothesis's or from anyone that will implement the forthcoming w3.org/annotation specification in the future, lets users commentary/community of their choice onto your content. Viewed through the lens of a Web extension implementation (vs a proxy) implementation this commentary can be applied without the tech ever needing to interact directly with your server. It's not being "pasted on top of your work" in the sense that other visitors who haven't made the same choice can't see it.

As a user of the web I have the ability to speak and listen freely (via twitter, reddit or other means). Those communities either do or don't do a good job at providing tools for countering abuse. Presumably, over the long term whether they do a good job has a direct impact on whether users continue to participate in them.

With annotation, the commentary is "on top of" your work in the same way that a tweet that mentions a URL to your website is "on top of" your work.

In fact, we have an open github issue describing a mechanism by which users could pull in tweets, or reddit or facebook posts about the pages that they're visiting automatically. For researchers, they could "listen" to Google Scholar to easily see the citations about the paper they were reading. You might call this "Persistent Ambient Search", i.e. the ability as an Internet user to electively be aware of conversations about the place you're at wherever you are.

Should publishers, bloggers, etc be able to prevent that from happening? Should twitter implement a "tweet.txt" feature that lets websites opt out of their URLs being mentioned in tweets? Should developers be prevented from shipping extensions that query twitter's API for tweets about the web pages a user is visiting?

We understand that the experience of annotation is palpably different, and that we're all envisioning a world in which it is essentially ubiquitous, but lets also understand the complex questions involved here as they relate to speech and the notion of what is in the public space. There is a middle ground.

Our community (we come together at conferences and many other venues, e.g. iannotate.org) has long discussed whether such a mechanism (call it annotate.txt) makes sense in the same way. There are several reasons why I personally don't think so. (I'm specifically discussing using robots.txt here, not the broader question of whether opt-outs make sense, which I agree with under specific conditions)

First, web crawlers consume resources. Anyone who's operated a website at scale knows that crawlers (from all of the search engines together) can sometimes represent a high percentage (in the double digits) of traffic. That's server load that the publisher has to pay for.

Second, search engines essentially republish content. The goal of indexing it is to expose it to others and make it discoverable in search results which usually are a foundation for advertising revenue for search services.

Annotation by contrast is speech. If you've published something publicly on the web then it's natural that others want to say something about it. They've already clicked on your web page and it's now inside their browser. When they comment on it, it's not consuming resources like a crawler.

If an opt-out mechanism is to be explored, it needs its own kind of signaling, and robots.txt is likely not the right place for that. With robots.txt you effectively "ban" crawlers from your pages, here I think the negotiation between you and The Public is more complicated. Page owners should be able to indicate a preference, but the public should also be able to override page owners' preferences when subject material is in the public interest.

Where that line is and how it becomes negotiated should be something we explore together.

This is going to be helpful on our local bike trail.

This. Is. A. Once. In. A. Lifetime. Moment.

What astonishes me is that my senator, Dianne Feinstein, is so consistently out of touch with where technology is headed, and so completely unable to frame her arguments in the context of why strong encryption exists in the first place. Further, there is no thoughtful exploration of the ramifications of her proposals, there's just the security state perspective and nothing else.

How about putting a public round-table discussion together to explore the issues?

Dianne, as your constituent as well as your neighbor-- one that lives just a mile away-- you are an utter disgrace as an elected representative to your people, particularly those of us in the technology industry.

This was cited by the NY Times as one of the brutally candid views offered by Ambassador Christopher Dell.

The relevant Wikileaks cable is here.

I think this probably creates a copyright issue.

Also, who would do the cloning, where would it be hosted. This would essentially fork content, and perhaps result in an even greater loss of control? What happens if the blogger wants to update the master, how do they update the copy?

I made this!

ho

Or... pretty much everything else. Human reaction time is nowhere near fast enough to dodge an incoming missile or other attack. I'm not a fan of killer robots, but this argument is incredibly disingenuous.

The counterpoint to this of course is that if annotation becomes as widespread, standardized and pervasive as we imagine it will-- and indeed hope it will-- then the notion that anyone can see the stall without the graffiti may become a moot point.

Can you believe it had maggots in it? Eewww!

test

test

We agree for essentially the same reason. That's why we formed Hypothes.is as a non-profit, according to a set of principles.

This is not completely accurate in two important ways. First, the Genius Web Annotator is a javascript client that can be injected either via a browser extension or via a proxy. Only the proxy version retrieves the page via their server. Their chrome extension does not. Our own open source annotation client at Hypothes.is works the same way.

Second, I think it's probably a stretch to say that a proxy redirect that is used solely for the purpose of injecting the javascript annotation layer is appropriation in exactly the same sense. Originally Rap Genius duplicated the lyrics on their website, depriving the original publisher of royalties or advertising revenue that they might have made. The proxy redirect doesn't permanently republish the work, but pulls it newly each time the proxy is clicked. Ads that are on the pages are redisplayed, paywalls are respected.

Certainly you could make the argument that because the proxy redirect strips the user information away and thus blinds the publisher to details about the user that might be helpful in targeting ads, that there is a possible commercial loss to publishers.

However, if you neither serve ads nor need meticulous analytics around exactly who is clicking on your site (probably most bloggers), then whether the reader comes via a proxied link or directly by browsing to your site may be more academic than substantive. The server load is the same and the end result is that your page is copied into their browser where they can read it.

Those of us in the annotation community (and particularly those of us that participate in the W3C Web Annotation working group) are working to deliver specifications and implementations of annotation that can eventually ship natively with browsers--eliminating the need for tricks like extensions or proxies. You'd sign into the annotation services (i.e. communities) that you want to listen to.

It's probably worth separating the technical implementation from the conceptual discussion of the speech that's happening (and in what ways it's both connected to and removed from what it references), since there are a variety of ways (some now and some in the future) that this layer can manifest. I'm not suggesting that there's no room for a discussion of proxies, or site owners' rights with respect to them, simply that perhaps there are two discussions worth having.

Is the corollary argument here that via a web extension I shouldn't be able to create an annotation on a remote server (even if that annotation isn't delivered and reanchored to the original content thru a proxy)?

Here's an extensive list we maintain (not exhaustively, nor accurately in many of the details, but as more of a scratch pad). We had extensive interviews and discussions with many on this list in developing Hypothes.is, including John Atcheson (mentioned below), who's a friend and colleague that I also worked with in assisting Getaround, as well as his co-founder Todd Herman.

The primary difference here is that the site is implementing, and integrating, the tech directly into the page. Medium is only "annotation" in the sense that it's in the margin-- in all other respects it's simply the same paradigm as the other flavors of web commenting systems, such as Disqus, Livefyre, fb comments, etc.

Web annotation in the sense of the W3C definition is a model where annotation services are provided by third parties and lie elsewhere. Users invite them to your page, and otherwise they're not visible to other readers who have not done the same.

Lets assume for the sake of argument that annotation providers did not offer proxy services, or that those proxy services could be prevented by publishers using something like an annotate.txt file (or even simply metadata in HTML headers that said effectively "please don't proxy").

But that even if you were explicit about not wanting your pages to be proxied, your pages could still be annotated by users with various browser extensions, or by the tech that is eventually likely to ship natively with browsers.

Would you have an objection to this kind of annotation?

Saying that the NYT lets you "select which articles are available for annotation" suggests that somehow they are exercising editorial control over annotation in the same way that they do for their native commenting platform-- which is disabled for selected and usually controversial articles, op-ed pieces, etc.

However, what the NYT did was demand that Genius disable their proxy service (what you get when you prepend a URL with genius.it/) for NYT articles. The reason that Genius complied is that the NYT could have blocked the proxy anyway (and may have done so).

However, NYT articles can still be annotated by using the Genius web extension. The NYT has no way of knowing whether the reader happens to deploy an annotation extension within the scope of their browser.

In this respect, Genius and Hypothesis function the same way. The architecture is dictated by the realities of the way the Web works.

Many of us feel that ultimately annotation tech may ship natively with browsers, eliminating most of the friction involved in their use, but also any recourse that publishers may have to block conversations. Thus the tension described elsewhere here between speaking truth to power and having control over conversations that are unwanted.

Your blog is still yours.

Because you're speaking publicly, people are going to talk about it, on twitter, on facebook, on reddit... and with annotation. There's nothing different about annotation except that it "magically" brings the things people say over top of your content. There's Genius, there's Hypothesis, and potentially someone may create a plugin that brings tweets over the top of the posts they're about. But the key difference here is that it's your readers that are consciously making the choice to bring those communities over your content. When I come to your blog post without those tools, your content is just the way you intended it, with only your clean unfettered page.

The problem of course is that if you provide a way to disable annotation, then it pretty much kills the "speak truth to power" angle.

This. So much of organizational conditioning and response-- and really modern life-- can be summed up in this one statement.

A fantastic insight that really gets to the root of different kinds of social psychology. Some people want to participate in and contribute to society. Some want to tear it down. Others want to poke it and see how it responds while sipping tequila.

It's the law!

The new Ben-Hur can't possibly beat the old one.

What's wrong with a little arsenic in your tea?

Actually the VA gets most sjdflajdf;kasdjf;ljasd.m

are developing

are

... into the foreground either as private, group or public comments.

s, N

... thus use cases in Europe will benefit from this.

Hypothes.is

"... implementing the emerging W3C standard open annotation data model ..."

Not sure what you mean by angles

Page

This looks interesting.

Sanders is never going to make it.

I wish more American companies-- well, actually just more companies period-- had the balls to do this. Apple has a lot of faults, but it needs to be said that very few companies would ever take this kind of public stand about anything.

Is there some particular reason this would take 5G speeds? Sounds like a 2G bandwidth event.

A screenshot of SideWiki.

Here's an example of their UI. Question: Can we improve upon the sidebar design approach? So many projects use it, ours included.

Here's a spreadsheet of such projects we've been maintaining. Edits welcome.

One of my all time favorites:

https://www.youtube.com/watch?v=k0t0EW6z8a0

here

Bpah

Here

This is the thing. https://help.passpack.com/knowledgebase/idx.php/46/107/article/How-to-Activate-Secure-Collaboration-Features.html

Perhaps I should set up such a home. Might help defray the monthly food bill.

Pardon me from asking the obvious-- but why don't people eat these "abandoned" chickens?

Apparently this might not be such a great idea

Dave Winer's thoughts on why Anywhere but Medium

Listen... here @ 17:29:

https://www.youtube.com/watch?v=te8AIbQ7tDg&feature=youtu.be&t=17m29s

I think we take this kind of BS as fairly pedestrian unfortunately. "Oh yeah, they said we needed to file a FOIA request." But think about that. Here is a high profile court case that depended on a rare lab technique that was created de novo for this one request, based on a single 1997 study (the info that this test was done uniquely for this one request was in the Netflix documentary).

Folks have questions about the procedure, and instead of being transparent about what they did, our government essentially says "Fuck you, force us to give it you if you really want it."

At least we have the docs here, so I guess we should be grateful for that?

Citation please? I've heard no serious arguments towards this end. He'll keep trying and eventually he'll succeed. It may take another year of failures. They're being paid full price for these launches, so from a traditional space services perspective, any experiments he wants to do to recover rocket stages at the end of these launches is really up to him. Who are these theoretical detractors? Strawmen for the purposes of writing clickbait?

Ha.. was this a deliberate teacher joke? If so, I applaud! http://public.wsu.edu/~brians/errors/chalkfull.html

I can see a perspective on this-- particularly in large organizations where individuals may primarily toil for the upward accumulation of wealth towards executives and shareholders.

But there is another perspective-- that of the individual scientist, the small entrepreneur, the social worker. Dedicated, passionate individuals performing magic by accomplishing things that others cannot, often unreasonably constrained by time, money and resources.

What about negotiating our rights away in secret, and subordinating sovereign rights to corporate interests? http://www.truthdig.com/report/item/heres_why_the_trans-pacific_partnership_agreement_just_plain_wrong_20150107

I hope the irony in pairing these sentences was intended.

Disagree on the 2nd point. Terrorist networks have vastly less impact on our citizens than many many other things, including citizens shot by police.

Somewhere Trump is eating an Oreo

Not to rain on this parade, but how much of this is due to unprecedented amounts of quantitative easing? http://www.startribune.com/unease-with-quantitative-easing-is-going-global/363399171/

And let's not forget Prescott Bush, father of George H. W. Bush, our 41st president. http://www.theguardian.com/world/2004/sep/25/usa.secondworldwar

As much as I dislike Trump, this "ritual" is essentially just common sense. If I were running for office, I'd do exactly the same thing. Why deprive yourself of this intel? And the video tapes of him succeeding? Could just be a homegrown form of positive reinforcement that he uses to review his interactions and refine his messaging.

Thanks for the coverage David, and I absolutely love the OA search engine you've created with JURN. If you read this, reach out to connect with us, I'd love to chat.

We've got our own list too.

Note that it reads "much of scholarship". Even without the majors (though they're already curious), I think we'll get there. What this really requires is momentum and buy-in from users and a majority of publishers, but not necessarily all of them. The browser extensions already give us a way to reach into content we don't control-- but support from major publishers helps us with content that's more challenging, and helps to signal to readers a default solution.

These are great questions, lets break them down:

Hand annotation-- I think this one is easy. People will annotate what they want to discuss, i.e. people's attention will be spent as they determine it should. Clearly not everything will get an annotation-- but then there are many papers that are never even read. However, the more pervasive, standards-based, powerful and user-centric an annotation capability is, the more it will be used. And by that I don't mean the Hypothes.is service, but rather a client through which you can connect with your annotation service of choice.

Linkrot checking and repair -- This could easily be assisted through automation. I can imagine us crawling annotations periodically (and/or checking them when they're served). Broken links could either be signaled as an issue to the original annotator, or potentially checked against a web archive (e.g. archive.org) and supplanted with a reference there-- or both.

Moderation-- Obviously this is the most important question. If annotation becomes spammy, then users and (particularly) publishers will stop using it. But there are wide variety of strategies we can explore to address this, including community flagging, leveraging groups, using algorithms to identify likely spam and focus resources on it, gating initial permissions by scoring registration email addresses and other such options.

We will shortly support annotation of PDFs against their Adobe ID fingerprint. This will remove the need for a stable URL for PDFs, and will enable use cases like annotating PDFs that you save to disk or receive via email from others.

Otherwise we can also relate two HTML pages to eachother (or to a PDF) if the same content is available elsewhere.

The video I think they're talking about is here, though at the key point the camera points away. https://www.youtube.com/watch?v=cX5CPx4RKWw

This: http://www.hoverboard360.com/

Not to be confused with this: https://hoverboard.com/

Is the money Taboola offers really worth the degree to which it debases the content it's embedded on? (Don't answer that!)

Or perhaps, "contribute annotations"?!

But a bunch of annotations ^

We also support embedding the javascript application on a page, so that it's there by default. https://hypothes.is/for-publishers/

Blah.

What? No one should ever try to redesign others' work because they might have less than complete information about the decisions that went into it? "cough" Bullshit!

Like this.

Here's LaTeX math support in the pull quote.

$$\(a^2+b^2 =(p^2-q^2)^2+(2pq)^2 =p^4-2p^2q^2+q^4+4p^2q^2 =p^4+2p^2q^2+q^4 =(p^2+q^2)^2 =c^2\)$$

$$1/(1-x)=\sum_{n=0}^\infty x^n$$

$$\begin{matrix} -2 & 1 & 0 & 0 & \cdots & 0 \\ 1 & -2 & 1 & 0 & \cdots & 0 \\ 0 & 1 & -2 & 1 & \cdots & 0 \\ 0 & 0 & 1 & -2 & \ddots & \vdots \\ \vdots & \vdots & \vdots & \ddots & \ddots & 1 \\ 0 & 0 & 0 & \cdots & 1 & -2 \end{matrix}$$

Saying relevant.

How exactly does one reflect ... strategically?

Add another note.

Write something here that I feel passionately about.

مل المسار الانتخابي مستقبلا، واستخلص بأن هيئة تختصر صلاحياتها في المراقبة الحصرية للانتخابات كما ورد في الرسالة الرئاسية الأخيرة، لا تعبر سوى عن إرادة الإطالة في عمر التزوير المتفشي في المنظومة السياسية الوطنية بأشكال ووسائل أخرى.

Hi!

Hello, I really don't think so.

Here.

This area here.

Here.

Make an annotation

If the battlefront on climate change turns to whether extraction industries have misled investors, that will be an extraordinary turn of events indeed.

Here.

Like this.

This is a test message from Dan

test

Like so

Say something appropriate here.

Test

It would have really been a service to readers here if you could have explored this a bit. What exactly are the limitations of flat discussions Jeff?

I'll take the liberty of offering my list:

Flat discussions:

1. Don't scale well to many participants. After a very short while the conversation maxes out, and people are talking over each other.
2. Don't work well on popular posts. This is a variation of the first point, with the emphasis on participants in a compressed period of time.
3. Force hacks like the use of @mentions at the beginning of lines to target remarks to a specific person.
4. Get noisy quickly.
5. Dumb down conversations to the few ideas and few voices that are sustainable before things get too congested.

Actually only in older usenet style systems. In forums like Reddit, you don't have to click through anywhere, and the 'collapsor' at the root of any branch lets you quickly "read-by-collapsing' as you quickly decide that each branch doesn't deal with a subject you're interested in.

In other words, having the conversation you want to have is "getting off topic". Says who? Who decides which conversation is "on topic"? The tyranny of the single loud room, in which only simple conversations can be had by a few folks?

Not to belabor the point, but the whole point of threading is to allow people to "start a new topic" if they want to "get off topic", but without disturbing those who prefer to continue discussing the thing they were.

Often so are normal discussions. Does that mean we shouldn't have them? Conversations are complex things.

test

Something that I'd like to say.

I think stickiness should be for top-level source annotations, not replies.  The reply should default to the parent, particularly for groups.

How did you do that?

Agree with this. Wider is better.

VERY soon.

I actually don't see it that way. When I ask if "that makes sense", I can just as often if not more likely be asking whether I make sense-- appealing to them as the sensible judge, and implying that perhaps I am the one that is "non-sense".

Someone who is perhaps overly self-critical might see it the other way around of course, but I don't think the question as stated carries a negative connotation.

Test

Wait, not a whisper in this entire article about what utter crap the F-35 apparently is, completely unsuited for many of the fundamental tasks it was designed? The cynic in me is amused that the Chinese may have replicated a turd, the counterpoint of course is that they may have been following the news and fixed the flaws themselves.

The first Beyond the PDF workshop was a grass roots affair, held at UCSD in 2011. An except from the page states:

The goal of the workshop was not to produce a white paper! Rather it was to identify a set of requirements, and a group of willing participants to develop a mandate, open source code and a set of deliverables to be used by scholars to accelerate data and knowledge sharing and discovery . Our starting point, and the only prerequisite to participating, was the belief that we need to move Beyond the PDF (meant to capture a common philosophy, not necessarily to be taken literally). In a heady moment we might also describe our efforts as the desire to contribute to the development of a free and open digital printing press for the 21st century.

Haven't you just contradicted the false dilemma you posed in the first paragraph? I was just about to leave a note there venting a little frustration, but here you've knocked that straw man down. Perhaps recraft the opener instead?

test

Richard Stallman is famous for the speaking rider he sends out when he is considering an engagement. The relevant portion:

If you can find a host for me that has a friendly parrot, I will be very very glad. If you can find someone who has a friendly parrot I can visit with, that will be nice too.

DON'T buy a parrot figuring that it will be a fun surprise for me. To acquire a parrot is a major decision: it is likely to outlive you. If you don't know how to treat the parrot, it could be emotionally scarred and spend many decades feeling frightened and unhappy. If you buy a captured wild parrot, you will promote a cruel and devastating practice, and the parrot will be emotionally scarred before you get it. Meeting that sad animal is not an agreeable surprise.

(Yes, I know that's a cockatiel)

I'll give a separate answer, which is that I think there are probably many different models, and hopefully what emerges over time is a spectrum that varies according to context, classroom, strategy, etc.