Initial meetup and discussion

I have reserved room 529, a large classroom, on the fifth floor of USF's downtown campus at 101 Howard at Spear street:

http://www.usfca.edu/downtown/visit/

I think after 6 o'clock the meters become free and there is likely street parking but also garages. BART is only two blocks away at the Embarcadero station.

Note: the downstairs door locks at 7:30 sharp! You will not be able to get into the meeting if you come after 7:30.

TOPICS:

• parrt: Parsing state-of-the-art summary, ANTLR 4 (plugin) demo

• Oliver Zeigermann: TypeScript

• parrt: Foodfight concerning (explicit) static versus dynamic typing. Please come prepared with cogent arguments.

I will moderate and act as control rod in the nuclear reactor for all discussions.  This is meant to be a social and pleasant experience so behave yourselves. :)

Looking forward to meeting everyone and having a great conversation!

Join or login to comment.

  • Gerald R.

    Excellent

    March 7

  • Terence P.

    Oh, technical report on ALL(*) parsing with MUCH empirical data. http://www.antlr.org/papers/all...­

    March 7

  • Terence P.

    Loved it! Thanks so much for everyone that made it :)

    March 7

    • Warren S.

      Quite enjoyable. Am checking out Julia, as a result. Meanwhile, slides? In particular, I would like to show the Antlr4 metrics to some colleagues.

      March 7

    • Terence P.

      posted 1 minute after yours ;)

      March 7

  • Sergei W.

    I was wondering... Is there empirical evidence that programmers actually write tests to check for incorrectly typed method names and variable names, in languages such as JavaScript / ActionScript / Python where any object can have any method created at run time?

    Quite recently I was bitten by a Javascript blunder where I said something like

    element.innerHTML = "<p>error message</p>"

    - but this did not show the error message. Then I wrote a test that checked that element.innerHTML was actually being assigned the right value; and it was! But the error message was still not shown.

    March 7

    • Bryan H.

      I have found that, in practice, good web devs just don't make many typos. For most of my programming life, I did C/C++, but I've been doing JavaScript development for the last 3 years. At first I made typos all the time, and I realized it's because I got used to using C/C++'s static type system/IDE auto-completion as a crutch. Without that crutch though, I have found that the accuracy of my coding has increased tremendously.

      This isn't to say that these bugs don't crop up, but I have found it's just not really an issue "in the trenches." More iterative/testing centric approaches to development combined with code reviews (which everyone should be doing anyways IMO) tends to mitigate the downsides to dynamic typing reasonably well.

      March 7

    • Warren S.

      An alternative to explicit type, Auto-completion can be derived from namespace. Trivial example: I have been editing an Antlr3 grammar in Xcode, with the file type JSON, so that it will auto-suggest words already declared in that namespace. Or perhaps namespace could be considered the type. Then auto-complete may be combination of backtracking the naming taxonomy while minimizing the hamming distance.

      March 7

  • Oliver Z.

    Link to the TypeScript code samples: https://github.com/DJCordhose/ty...­

    March 7

  • Sergei W.

    Here is an example of where a type system is perceived to be "in the way". Suppose we have a function getOwner() that takes an Animal and returns an Owner. Now, we could have derived classes, such as Dog and DogOwner. We "know" that getOwner(Dog) "always" returns a DogOwner. However, the type system of Java is not strong enough to express this guarantee. Therefore, we must downcast getOwner(dog) to DogOwner by hand, sacrificing static type safety.

    Dog d = ...;
    DogOwner x = (DogOwner)getOwner(d);

    Question: What kind of type system, in what language, can express the guarantee that getOwner: Animal->Owner actually always returns a DogOwner for a Dog argument?

    March 7

  • Adrian K.

    As I mentioned, at our meetup San Francisco Types, Theorems, and Programming Languages (http://www.meetup.com/SF-Types-...­), we'll be working our way through Benjamin Pierce's "Software Foundations", which covers proving theorems with Coq. We're doing the first chapter for this coming Monday (the 10th of March), so there's still time to get in on the ground floor.

    1 · March 6

  • George S.

    Most interactive one I've attended. Look forward to more.

    March 6

  • Hans U.

    Any chance of recording the meeting and putting it up? I won't be able to make it but still have a desire to see the result.

    March 6

    • Terence P.

      Hi Hans, sorry. I'll post slides later. I should do one in the valley I guess

      March 6

  • Leo M.

    If someone wants to prep some Science for the food fight, check out the empirical research by Stefan Hanenberg: http://www.informatik.uni-trier...­

    March 6

  • Rick

    Alas, I won't be making it. Too much I gotta do down here, and the trek up to the City is not insignificant. I'd sure love to have a meetup down in SV, though. Yahoo might be willing to let it happen in their cafeteria. Or Hackerdojo, maybe (that's close to where I work).

    2 · March 6

  • Michael Robert S.

    Is anyone driving up from near Mountain View that I could join in a carpool? My roommate will likely be coming too and could also use a ride. I/we can take the Caltrain back too if it's a one-way ride.

    March 5

    • Brandon

      I'll still be leaving from Mountain View if others need a ride. Let me know and we can coordinate.

      March 6

    • Brandon

      I'll be leaving from Grant Road Shopping Plaza: 1250 Grant Rd, Mountain View, CA 94040 around 5:30. Contact me if you want to join the carpool.

      March 6

  • Terence P.

    Don't forget that the door downstairs and even the elevator will not let you upstairs after 7:30. (Also don't go out for a smoky treat... you will get very lonely downstairs by yourself.)

    Also, unfortunately, I cannot bring beer tonight as university has fun rules about requiring a security guard and a bartender etc.

    March 6

  • Terence P.

    Getting very excited about the meeting tonight! Those of you that are religious about static versus dynamic types, come prepared with your best arguments. Ollie and I will have our carnivore proof suits on. I am particularly interested in arguments against static typing; recall that static typing does not necessarily imply explicit typing. For example, a compiler can infer the static type you here:

    var x = 3; // duh, it's an integer

    Given the broad backgrounds of the participants, my state-of-the-art parsing will end up being a quick summary for the experts / speed demo and then a live programming demo where I build a grammar for simple programming language and then perform actions based upon listener events from the parse tree.

    1 · March 6

  • Cristina R.

    Jsp

    March 6

  • Rohit S.

    Hi

    March 5

  • Terence P.

    I think I'm going to start a foodfight concerning static vs dynamic typing and then propose a compromise.

    2 · February 20

    • Christopher B.

      Sorry I won't be able to make the inaugural meeting and hear the compromise first-hand. Thanks to Adrian King, I just discovered some 2000s work on hybrid type systems, which looks fascinating. The Sage language apparently lets you write a dependent type, then have the compiler automatically generate runtime checks for any of its declared properties that it can't verify statically, instead of always requiring the programmer to provide a proof.

      February 25

    • Terence P.

      Interesting. Thanks for the pointer.

      February 26

  • Reay D.

    Food fight sounds fun. Any chance we could meet in the South Bay for some of these? I really can not drive to SF for all of these.

    February 20

    • Christopher B.

      Also like this idea. Live and work in South Bay.

      February 21

    • Hans U.

      I would love this ideas as well. I'd offer my offices auditorium(Palo Alto) but we would have to pay for Security and Janitors.

      February 21

  • Terence P.

    I think a brief "state of the art in parsing tools and technology" would also be good to get this group rolling in terms of implementation

    1 · February 20

  • Dan B.

    I was hoping that sentence was just going to end at "foodfight."

    1 · February 20

    • Terence P.

      Well, i will try to bring beer, so...

      February 20

  • Oliver Z.

    Yeah, could do talk on TypeScript type system and why it looks so weird. Anyone interested?

    February 14

  • Terence P.

    I was able to reserve a big room at USF's downtown campus! It's the brick building on the corner of Howard and Spear, down the street from Google and Mozilla. I'll work out details concerning the front door status etc...

    2 · February 18

  • Terence P.

    We have a human team working on C++ output :)

    2 · February 16

    • Warren S.

      Ah yes, those Humans come in handy, sometimes.

      February 17

  • Sachinrao P.

    Would loooove to attend , but out of Cali till March 17th
    :-(

    February 17

  • Chris W.

    I'm very interested in talking about problems with current generation languages, but unfortunately I have a conflict with this date. Perhaps another time.

    February 16

  • Warren S.

    I am very curious about the state of Antlr C/C++. Very curious, as in meeting a deadline.

    1 · February 16

  • Terence P.

    Concerning locale: i can reserve something for sure main campus USF (near panhandle, away from bart) or possibly downtown campus but doors lock at 6 so a hassle. Any of you fancy industry humans wanna offer to host? (near decent parking or muni/bart?)

    1 · February 14

People in this
Meetup are also in:

Create your own Meetup Group

Get started Learn more
Rafaël

We just grab a coffee and speak French. Some people have been coming every week for months... it creates a kind of warmth to the group.

Rafaël, started French Conversation Group

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy