Hacker News new | past | comments | ask | show | jobs | submit login
Things you didn't know about Python (speakerdeck.com)
181 points by d0ugal on Oct 5, 2012 | hide | past | favorite | 41 comments



Do these postings of slides ever help anyone? I never seem to get anything out of just clicking through slides with no explanations. It's essentially reading 20 unjustified bullet points to me.


I agree. The problem is that modern slide design says that slides shouldn't be done in a way that lets them work independently of the presentation. That is, slides are there to provide structure and illustrate things that are difficult to describe verbally, but aren't meant to repeat what the presenter is saying.

It would be very nice if these online slide presentations came with the speaker's notes that actually cover the main points of what they're saying (with explanation). This assumes that these notes exist, but I feel that the slides by themselves are not very valuable.

(No disrespect meant to Armin Ronacher — I use his work every day and I'm a big fan of what he's done for the Python community.)


Really, the problem is that a live presentation and an online slide deck all by itself are two very different things, and trying to make a single slide deck that works for both is pretty much doomed for failure.


That would be more of an essay - which would certainly be valuable, but takes a considerable amount of time.


I've said it before elsewhere, but for me reading the slides to a presentation I haven't seen is sort of like watching tv in a different language. I mostly know what's going on, and I can follow, but I'm pretty sure I'm missing the main points.

That being said, seeing these presentation decks has often piqued my interest about some concept or other, and prompted me to learn it on my own. shrug


My personal take on that is that there should be a point in actually attending a conference versus just looking at the slides :-)


Then let's not spam the slides out to non-attendees.


I, a non-attendee, posted it to hackernews because I found bits of it looked interesting and relevant. Sure, I'd rather see the talk and I didn't expect it to float around on hackernews for this long but it shows there was some interest. I'm sure you don't find everything relevant and interesting on hackernews normally do you?


I enjoyed it. As a non-professional tinker, knowing even basic words and concepts (e.g. Twisted Python used for networking, something I didn't know before) gives me a key where to look.

Not all slide dumps are useful, but this was interesting to me.


Sure that it would add value to have the speaker's note but I found it actually helpful.

The duck typing part interested me as it seems a good fit for the problem I'm trying to solve: I have to work with a bunch of objects retrieved from a key/value store, some keys are simple attribute but others define the object class, required keys, value constraints, etc. while keeping the code generic. Are there any good references on this ?


the more extreme code talks tend to have really great slides - in this case slides are a better format than a blog post really because you can direct the audience's attention through the important bits of the code with your slide transitions, as opposed to just throwing gists at them. here's an example from a talk i just gave, the slides came out well imo[1], its basically the material's i'd use to teach my colleagues this stuff.

[1] http://www.dustingetz.com/2012/09/24/StrangeLoop2012-monads-...

really i think it comes down to, for a lot of talks, slides aren't value-add to the audience, they're just a loose outline for the speaker


This was clearly posted to rabble rouse and poke a finger in the eye of people who've been criticizing the language on HN; see slide 99. Classy!


>Do these postings of slides ever help anyone? I never seem to get anything out of just clicking through slides with no explanations.

Huh? I picked at least 5 different points and techniques I can use from this specific slides, some of them with code.


Armin's implementation of `cached_property` is not entirely correct. Well, it works, but the branch where `value` is not `missing` is never executed: the object's `__dict__` takes precedence over the descriptor as long as the descriptor does not define `__set__` method.

Here is an implementation of `cached_property` I use:

    class cached_property(object):

        def __init__(self, fget):
            self.fget = fget
            self.__name__ = fget.__name__
            self.__module__ = fget.__module__
            self.__doc__ = fget.__doc__

        def __get__(self, obj, objtype=None):
            if obj is None:
                return self
            value = self.fget(obj)
            # For a non-data descriptor (`__set__` is not defined),
            # `__dict__` takes precedence.
            obj.__dict__[self.__name__] = value
            return value


Don't do this. Your implementation does not work if someone invokes the dscriptor's __get__ by hand which is not uncommon. My implementation takes the shortcut but also still does the correct thing if you keep a reference to the property object.


Best slide for me was "worry less & get stuff done". Good advice.


Saw the talk in-person at Pycon South Africa this morning. Excellent presentation - thanks Armin.


Is there a video of this talk somewhere?


The talks are being recorded. They should be online later. Great talk from Armin!


Best I can find is https://www.youtube.com/watch?v=Ug8jBO6QMaY which has the audio (low quality w/ buzzing noise), but video is just a blank green screen the whole time.


some context for slide 99? (which says "screw HN")


It was a tongue-in-cheek reference to the recent discussions on HN about Python being in decline.


on HN every technology older than a week who isn't posted by very vocal 19 year old fans at least 4 times per day is 'dying'.


Can we please try and avoid the casual ageism? I'm 20 and I don't recognise my 19 year old self in that description. Further I know many people of all sorts of ages who switch to the latest 'cool' technology at the first opportunity.


The poster didn't say "every 19 year old is a vocal fan boy who posts 4 times a day"

The poster said that there are some 19 year olds who are very vocal about their fanaticism and who post 4 times a day.

It's not ageism because it's not generalized to an entire group of people. So if you don't recognize yourself in the comment, it's not aimed at you. Stop being hyper-defensive.


To imply being 19 or age in general has anything to do with being vocal or a fan is agism. What makes zalew believe those posts where made by 19 year olds? Last time I checked, HN did not require age verification or disclosed such information about its members.


We're quickly approaching that insane point in this discussion where we're now forensically dissecting what appears, to me at least, to be a throwaway comment by zalew. If zalew meant the comment to be a deep critique about the intersection between age and vocal fan boy-ism then I've missed that.

I don't see the comment as ageism. I see the comment as constructing a strawman in order to point out that the more vocal proponents a new technology has, the more attention and interest it can draw on a site like hackernews. It's a subset of the problem that on the internet, in front of an anonymous audience, points of view (or arguments) that are forcefully presented and appear to be logically consistent can often beat out points of view (or arguments) that are less well reasoned and less forcefully presented, even if they are more factually correct.

Thus 'python is dying' becomes a thing because it doesn't have the same level of noise surrounding it as other technologies. The lack of noise doesn't mean python is dying, it means python is less noisy (whatever that actually means).

But to get back to the point I was trying to make. The comment doesn't really say anything about being 19. Sure attach an 'immature' or a 'lacks experience' and now we're trending towards stereotypes that could imply ageism. Same as if zalew had said "a 56 year old condescending, out of touch, past their prime, vocal grandfather of two, great grandfather of 6, fan boy."

But simply sticking 19 next to some other words doesn't automatically imply ageism. And anyway I was reacting to ElliotH's need to defend himself from the characterization made in the strawman argument. If you want to go around fighting all things ageism that's your right and you're welcome to do it. But to spend time fighting strawmen, well, that seems a bit ridiculous. Although, at this point no more ridiculous than the amount of time and effort I've put into this reply. So I guess we're all suffering from the same thing this morning. Must be something in the company coffee.


Or maybe, just maybe, he was being sarcastic and not ageist? God, why do people need to take everything at face value...


I agree with you for this particular case but you are mistaken if you think that sarcastic and ist are mutually exclusive. Sarcastic comments of a ist form can (and nontrivially often do): mask sincere ism; be a more socially-acceptable-looking way of expressing actual ism; encourage ism, especially of a casual degree; exhibit societal/subconscious/not-actively-malicious ism; or create a generally hostile environment for people of an oppressed class of *.

So, yeah, probably an overreaction for this particular occasion, but it's not so easy as writing it off as "sarcasm".


>To imply being 19 or age in general has anything to do with being vocal or a fan is agism.

It's also true.

>What makes zalew believe those posts where made by 19 year olds? Last time I checked, HN did not require age verification or disclosed such information about its members.

Most of the pages linked touting new technologies have an "about" page or a bio of the author.

Also, older devs have enough years of experience with other people --on the job, on the web, on conferences etc--, to know that it's mostly young developers that go crazy over such things.


"19 year olds are fanatics" !=> "fanatics are 19 year olds"


Except Lisp...that's what hackers use or something :P


There was a recent story here claiming python was in decline, guessing it's that?


He might be also alluding to recent (~few months) influx of Rails and Node.js related stories.


It looks very interesting, but I feel I'm missing half by seeing just the slides. Is there a video of this talk somewhere?


Hm, I see on the schedule page of PyCon-ZA that the talk was only this morning? (or night, depending on your time zones) So I can imagine that if there'll be a video available, we'll have to wait for it :)


If you are interested in learning more about python and caching I wrote a post about it couple days ago:

http://russell.ballestrini.net/explaining-cache-with-python/


That's more accurately described as memoization, as caching is a broader term.

Also, you might want to look into Sieve of Eratosthenes, as you only have to test up to sqrt(n) for primality: https://en.wikipedia.org/wiki/Sieve_of_Eratosthenes


Thank you, AncientPC. I have adjusted the article with the insight you and many others have provided me.


I have a ton of respect for Armin. Can't wait for this video to be posted.


I especially liked the last slide :)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: