A test of data annotation in an IPython notebook.

A test of code annotation in an IPython notebook.

Or, perhaps, "teaching humans /and/ teaching machines"

Kasparov's Equation: (weak human + machine + better process) > (strong human + machine + inferior process).

"The winner was revealed to be not a grandmaster with a state-of-the-art PC but a pair of amateur American chess players using three computers at the same time. "

https://www.palantir.com/2010/03/friction-in-human-computer-symbiosis-kasparov-on-chess/

The "precariat"

In one of those cycles -- though probably not this one -- it won't be speculation and it won't be the future.

Considering the candidates on this stage did not make the main event due to low poll numbers, whether or not they can actually go the distance is an important concern.

http://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FzgnHF2CwrPs%3Ffeature%3Doembed&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DzgnHF2CwrPs&image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FzgnHF2CwrPs%2Fhqdefault.jpg&p=1&key=e898f7f22a3443a6893439149a6fb36f&type=text%2Fhtml&schema=youtube#

As a non-controversial, non political figure, Fiorina has a great amount of potential coming into the debate to break out of the pack, and gain some name recognition on the national stage. Which, she ended up achieving, with most major news networks reporting the event as her win.

http://www.cnn.com/2015/08/06/politics/carly-fiorina-republican-debate/

She also has a personal story regarding her rise from secretary to CEO of Hewlett Packard that is not only appealing in and of itself, but it makes her seem a staggeringly different kind of business owner than one Donald Trump, who she addressed quite intelligently later on. Quite simply, Fiorina has much more street cred with small business owners.

However, the performance on stage was not particularly stellar, so her superior showing did not require any particularly outstanding moments. Nevertheless, she’s one of my favorite candidates in the coming election cycle, and definitely one to watch.

I was particularly impressed by her responses to the foreign policy questions in the debate, as well as the below post-debate interviews. I may not entirely agree with all of her stances, but the ones she took ranked high for consistency and to an extent, feasibility.

http://www.cnn.com/2015/08/07/politics/carly-fiorina-donald-trump-name-calling/

With a total of three gubernatorial terms under his belt, Perry ran for the presidency in the 2012 election cycle, only to drop out shortly after an extremely poor showing in one of the debates.

http://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FkTNjhcyx7dM%3Ffeature%3Doembed&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DkTNjhcyx7dM&image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkTNjhcyx7dM%2Fhqdefault.jpg&p=1&key=e898f7f22a3443a6893439149a6fb36f&type=text%2Fhtml&schema=youtube#

Returning to the stage, with a new pair of glasses to boot, the governor needs to work to redeem himself in the eyes of the public as a viable, consistent, candidate.

Near the threshold for fault slip and earthquakes

"Juan Zarate is an idiot" says Gary McGraw, https://twitter.com/cigitalgem/status/628705116049276928"

Again.

Observation: significant tissue changes
AnalysisMethod: vbm 
Location: substantia nigra
SubjectGroups: FTL compared to CTL

Observation: significant tissue changes
AnalysisMethod: vbm 
Location: midbrain
SubjectGroups: FTL compared to CTL

Observation: significant tissue changes
AnalysisMethod: vbm 
Location: dentate
SubjectGroups: FTL compared to CTL

Observation: decrease
AnalysisMethod: vbm 
Location: cerebellum
SubjectGroups: FTL compared to CTL

Our local spot, Toad in the Hole, gets this right with the help of a thoughtful touch I wish more restaurants would figure out.

In this photo you can see two acoustic baffles hung high. There are more all around the ceiling. So simple, so effective.

Update: "It is always quiet enough to talk"

Observation: no difference
AnalysisMethod: vbr
Location: all brain
SubjectGroups: FTL compared to CTL

Observation: significant (p < .001)
AnalysisMethod: correlation
Location: iron deposition: Caudate head, UDRS
SubjectGroups: FTL

Observation: significant (p < .001)
AnalysisMethod: correlation
Location: cavitation: lateral globus pallidus, UDRS
SubjectGroups: FTL

Test annotation.

test

test

f the subject? whose phon, syn, and sem information? It could be interpreted as "all information of the sentence" which is surely not meant in this context.

The use of little nPs (paralell to Little vP) in some minimalist accounts makes possessives and prenominal genitives to subjects/external arguments, i.e. in probably exactly this kind of "more general sense" that you meant for HPSG

It is right that "Bären" is in the scope of "groß", but not necessarily that "Bär" means "a lie". "aufbinden" means "to tell something wrong to so.", "Bär" means then "a story". To some extent, one could say that the idiom is not "einen Bären aufbinden", but only "Bär", taking into account that the meaning of "aufbinden" does not change. That is, the modifier would have scope over the entire idiom.

This formulation is not so clear in comparison with the German counterpart (their variety >> author's?, theories? idioms?):

Oder vorsichtig formuliert: kein großes Problem, denn trivial ist die Behandlung von Idiomen in ihrer ganzen Buntheit nicht

Better??: "in all their many aspects?" instead of "in their variety",

It would be good to make this point clearer with an example.

"Referring to the examples would clarifies the term "self-embedding"

a sentence

What does "that-Verb" mean?

that-Verb?

I'm not sure if [Subj V [that-S]] and [that Subj V [that-S]] are \"the same construction\". Taken Goldbergs definition of construction, they are actually different in form and also in meaning (cf. illocutive force in embedded sentences). This point would be true if you have 4 clauses, then the 2 constructions in the middle really would have the same form AND meaning.

Or a sociopath for whom it's all just another day at the office.

And let's reward innovators who leverage and recombine mature technologies to solve real problems as richly as we reward those who create novel technologies that may solve no problem at all, or perhaps only the narrow problems of the tech elite.

That's a keeper.

That is what I've been looking for.

Good to know!

This was not in the New Yorker article.

Now I need to know the diff between recurrence segments A and D.

Peaks around 450. (Presumably years?)

Or this one?

This might be the series I am looking for.

This seems to be the series I'm looking for, although the New Yorker article talks about 41 datapoints and this is a series of 18.

Apps are important on phones because the web experience is dismal.

Other way round. Phones make the web experience dismal to push apps.

More information on resveratrol can be found at: CTD

My daughter, a geriatric social worker, sees this all the time. It's an amazing and wonderful thing.

If the geological data can tell us that number of events so precisely, I would guess -- but the article doesn't say -- that it also tells us with some precision about the regularity of recurrence. But then:

Recurrence intervals are averages, and averages are tricky: ten is the average of nine and eleven, but also of eighteen and two.

Which suggests that we don't have data on regularity of recurrence.

Which is the case? A crucial question that's frustratingly unresolved, and leaves me unable to evaluate the premise.

I'd love to see more context attached to this point.

The use of feeds described here doesn't strike me as a way for publishers to enable their visitors to annotate, but rather as a way to gather and present annotation activity. Perhaps call it out as such? And, in that case, the Atom use case is just a variant of the API use case which we talked about here

Shout out to my former team!

Is that proliferation a problem?

Collaborative curation/categorization is the way forward.

Is there a summary of prior predictions and outcomes?

Prediction for Horizon Report 2018: Seeing Spaces

Still some work to do on that front. The global conversation can be as enlightened or vicious as the village conversation.

https://www.linkedin.com/pub/davy-van-de-moere/1/a80/375

Filip Maertens

I'm in the same boat. I've never been a reader of e-books, but it's harder and harder to focus on any long form material.

It may be self-serving rationalization but I am lately exploring the idea that annotation of digital texts is a way to dial down distractions that take me away from them and dial up my engagement with them.

I think this would be a great use for what, in other systems, are called emoji. It might or might not make sense to offer the conventional ones in Hypothesis: thumbs up, happy face, etc. But I can see a clear case for teachers providing students with custom icons that -- as in this example -- denote Who/What//When/Where elements in a text. They'd be essentially a controlled vocabulary, on which search and filters could operate as with tags.

I think about this a lot. In general, reading online invites us to leave the text to explore external associations. Annotation takes us deeper into the text and will, I hope, help launch a slow reading movement.

This is a page note.

Private groups: always a top requirement.

Evidence for why a social component to Hypothesis matters in education.

See http://www.pressdemocrat.com/news/4151286-181/mark-kostielney-former-head-of

Can this expression be a target?

That was really disappointing, and it seems to have been the case for years.

Still true, still needs to happen.

Don't forget: also needs a custom-build extension for every major browser.

Yes! http://twitter.com/JoeGermuska is spot on. It'd take a decade to standardize dataframe but it'd be worth it.

Russell Beattie commenting on Tim Bray's https://www.tbray.org/ongoing/When/201x/2015/06/05/End-of-HTML

Tim has learned well how to write a clickbait Death-Of-X headling but of course he doesn't mean that, it's a nice overview, and it's attracted good comments from a number of good folks.

+1

Yes! That's why the Hypothesis model is so appealing for collaborative editing. You attach discussion to specific parts of a document, at any point in the workflow no matter how it's implemented. /cc @dwhly

Annotating from http://www.springbio.com/products/MMP-19/E403

Organism: Rabbit.

Annotating from https://scicrunch.org/resolver/RRID:AB_187910

Conversely patterns that start well can end well. A great example is MathOverflow. From When Open Access is the norm, how do scientists work together online?

*Another key success factor was the founders’ strong leadership in the early going. They set high expectations for the quality and tone of discourse. “We made it clear that you wouldn’t just ask an unmotivated question,” Morrison says. “You’d put a lot of thought into your question, you’d explain why you were asking, and you’d specify what would constitute a good answer.” Discourse was expected to adhere to the norms of professional math. “We had to say, look, you wouldn’t behave that way in a graduate seminar, and you won’t behave that way here.”

Today the site mostly runs itself. There are still moderators who “take out the trash” when needed, but the need rarely arises. That’s partly a reflection the StackExchange’s “wonderful support for community moderation,” Morrison says. But it’s also a validation of how the founders began with a clear statement of values and expectations.*

A related service: keep track of many addresses (and representations) of a given work.

http://t.co/cp7mhadfEA

Creating that option would force FB to assign a per-capita value to the surveillance. Awkward! And likely more than the market would bear.

https://groups.google.com/a/hypothes.is/d/topic/support/ZowpLBJp2F0/discussion

https://groups.google.com/a/hypothes.is/forum/#!topic/support/wQmnrKlgSqA because target URI truncates to https://groups.google.com/a/hypothes.is/forum/.

Is this different from self-directed learning?

Embodied cognition?

A kind of re-balancing? Appropriate this space to create something beyond managerialism?

‘the theoretical and practical knowledge of social life in the community’ (Lefebvre 1969: 155)

hmm....saturated? Is the LMS a teaching landscape? Is the open web another?

hmm....saturated? Is the LMS a teaching landscape? Is the open web another?

http://www.slideshare.net/mdvfunes/lived-time-life-beyond-clock-time

two sides of a coin that create the problem? In trying to solve it, perpetuates it? I have no idea what this means...yet seems important to understand it

Discuss?

I love this! teaching people how to inquire early

I have difficulty with this type of language.

Is this part of the 'why' rage should be situated?

In the UK we have: "We have rather lost control over the form, structure and function of academic knowledge; the determination of the times and spaces in which we teach and learn; the relationships between educational philosophies and the material environments of teaching; and relationships between students and teachers."

If this is so in this sweeping kind of way, then may be I need to become more of an activist...

It’s your classroom; occupy it with some important and creative lessons! Amen to that! and yet becoming harder to do with LMS requirements

but authors suggest caution:

"As Judith Butler advised, while acting out can give a buzz of empowerment, ‘it’s really important to be able to situate one’s rage and destitution in the context of a social movement’ (Bella 2011)"

I am not sure I have ever situated my rage in the context of a social movement. Should I? Why?

Well, people what might it mean?

We understand what is occupying a space and then adapt what is there for alternative purposes. It is how knowledge grows.

This reminds my of Jim Groom's idea of content being the residue of relationships that exist in space time...

The Really Open University love that!

I love the evocative words...and I do not really understand them... is it as simple as a curriculum can be a violent abstraction if it is fixed and 'unoccupied'?

Discuss please? Their lips move but I cannot hear what they are saying

This seems to imply an asserted need for embodied ( I prefer it to materialist) understanding?

And does this landscape now include the virtual? How?

I love the idea that you cannot evict an idea :)

Sounds familiar?

I am having difficulty with the overly generalised propaganda. This depersonalised the activity of education. Some educators might produce such beings but not all?

I never thought about this before. The very existence of something in space and time can be pedagogical? They know they will not succeed but they will teach?

How can we enable the production of autonomous subjectivities and liberating relationships within and beyond formally ‘occupied’ territories and environments? Is this not what the open education movement is trying to do?

Okay 'the violence of capitalist abstraction' somebody give me something beyond surface meaning?

Critical Practical Reflexivity has a certain ring to it

What does this mean?

I have been interested in Occupy for a while. What an interesting use of an explicit metaphorical mapping. Occupy as a non-physical activity

Connections to third space and heteropias?

This paper is not about online spaces, but resonates strongly with connected learning spaces?

http://www.tandfonline.com/doi/abs/10.1080/00335631509360475?journalCode=rqjs20

Indeed. What part of my sensorium is active when I receive signals reporting interaction with the presence that I project into online spaces?

And in the case of aboriginal songlines, places.

5, 1, 201

5, 1, 203

5, 1 187

5 , 1, 189

Not necessarily, if PDF is in the mix, it can be the medium for annotations that might later anchor to a website -- even if PDFs are distributed to participants and used locally as mentioned above.

How do we communicate with one another?

In so many ways.

I report -1 for the stream in the PDF case because https://hypothes.is/stream?q=uri:https:foo.com%2Fbar.pdf tries to load as a PDF.

Annotations anchor in the sidebar in both cases, but in 4.1.32 highlighting, and sync between highlights and the sidebar, is confused.

Just a timeout. Recheck.

This is a bug. With 4.1.32 some PDFs anchor annotations correctly in the sidebar but fail to highlight them correctly. TBD: Find out which PDFs and why.

This difference does not reproduce interactively.

Does not reproduce interactively. Check this on the next run, if persists, debug the test.

No diffs here, simply a case where the annotation target fell off the event horizon.

Underscores the (well-known) need to find a way to display -- or at least signal the existence of, and provide access to -- orphans.

Verified interactively: 4.1.32, while much faster (3.0.2027 is barely usable), only anchors 1 of 2 annotations.

The one that anchors has voluminous text. The one that does not is the single word re-test

https://hypothes.is/api/search?uri=https%3A%2F%2Ftwitter.com%2FKenNeumeister%2Fstatus%2F582190731592396800

Verified interactively. As per the other Medium example above, 4.1.32 can anchor annotations that 3.0.2027 cannot. Cool!

Verified interactively: $('.annotation').length == 9.

So this is case of a clear (but very minor!) difference in anchoring between 4.1.32 and 3.0.2027.

That such cases are rare and minor is remarkable!

Verified interactively: $('.annotation').length = 8

network or js timeout?

No diffs between 4.1.32 and 3.0.2027 here, but it's a mystery why no annotations show up.

Hmm.

https://hypothes.is/api/search?uri=http://www.bbc.co.uk/news/uk-scotland-glasgow-west-26243567 -> 4

But the app is using:

https://hypothes.is/api/search?limit=200&offset=0&order=asc&sort=created&uri=http:%2F%2Fwww.bbc.com%2Fnews%2Fuk-scotland-glasgow-west-26243567 -> 0

Why a different result for the URL-encoded variant? And why is this the only case I've found that exhibits the behavior?

I'm really curious about why the vastly different highlight counts here.

Not reproducible interactively.

document.getElementsByClassName('annotator-hl').length == 10

Evidently a test artifact. Explore why.

This annotation is an orphan:

                    {
                        "type": "TextQuoteSelector", 
                        "prefix": "nymous with faith in the future\u2014", 
                        "exact": "both in the future as a better world and as one in which the United States bestrides the globe as a colossus.", 
                        "suffix": "Technology has long been the un"
                    }, 

Presumably that's because it lives behind the paywall.

many annotations, slow loading, hence timing variability

This appears to say that even though both versions only managed to load 5 of 937 annotations before the timeout, 4.1.32 managed to draw a bunch of highlights whereas 3.0.2027.

Hard to confirm interactively because the larger issue is that, 4.1.32's superior performance notwithstanding, we are effectively unusable at this annotation density.

In theory, via in column A should not differ from its counterpart in column B. But slow-loading pages are vulnerable to variable network latency. I think that's why the difference here.

Still reflecting on what it means when highlight and bucket counts differ. Is 4.1.32 better than 3.0.2027? Worse? Or do they just differ in an unimportant way?

In this case, 4.1.32 gets right what 3.0.2027 does not.

http://en.wikipedia.org/wiki/GetThere was a search engine? Or did @dwhly not tell us about something else?

I report -1 for the stream in the case of PDFs because an URL like https://hypothes.is/stream?q=uri:http:%2F%2Fjonudell.net%2Ffoo.pdf tries to load foo.pdf.

timeout

This works interactively, check result of next test.

timeout

timeout

timeout

Probably timing, but compare to a future run.

timeout, try again

Very curious about the large # of highlights here!

timeout, try again

timeout, try again with > limit

With large #s of annotations, only some will load before the timeout. This time through that was 15 secs, I'll bump it next time.

Remember: for now this will never be > 20 because of inability to send PageDown or ScrollDown events using Chrome WebDriver.

Still exploring whether differences in highlight and bucket counts between 4.1.32 and 3.2.2027 mean that the former is more correct, less correct, or just different.

Only 1 annotation expected, but 2 found for blog.jonudell.net, however 2 expected and found for blog.jonudell.net/.

In this case the new code is more correct.

A /lot/ of time. But it's turning out to be well spent.

Step 11, part b: add a Hypothesis note to this sentence.

This is step 6.

This is step 3.