5 ms·

R.I.P. Phil Bagwell

182 points by SanderMak 14y ago

rossjudson 14y ago
See http://infoscience.epfl.ch/record/64410/files/techlists.pdf http://infoscience.epfl.ch/record/64410/files/techlists.pdf for one of Phil's most influential papers. Surprising (and saddening) to hear of his passing. See also http://lampwww.epfl.ch/~prokopec/ctries-snapshot.pdf http://lampwww.epfl.ch/~prokopec/ctries-snapshot.pdf for a more recent contribution.
- marshray 14y ago
  Wow, this Ctrie looks like a concurrent Swiss Army data structure. Very interesting, thanks.
fogus 14y ago
Some background on the kind of person Phil Bagwell was and his contributions http://blog.fogus.me/2012/10/15/phil-bagwell-rest-in-peace http://blog.fogus.me/2012/10/15/phil-bagwell-rest-in-peace Of particular note is his work creating Hash Array Mapped Tries (aka functional associative arrays) detailed in http://lampwww.epfl.ch/papers/idealhashtrees.pdf http://lampwww.epfl.ch/papers/idealhashtrees.pdf and implemented in Haskell, Clojure, Scala and Rubinius. His latest work involved lock-free versions called CTrees[1] that are likely to create more waves as others become familiar with them. [1]: https://github.com/axel22/Ctries https://github.com/axel22/Ctries
- carterschonwald 14y ago
  is there a good spot to find all the papers he's coauthored?
  - SanderMak 14y ago
    Here's a start: http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/b/Bagwell:Phil.html http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/b/... doesn't include technical reports though.
noelwelsh 14y ago
I only met Phil once, but it was his suggestion that led to the formation of Underscore, where I now work. I hoped I'd get a chance to say thanks, but now that opportunity is gone.
- berserkpi 14y ago
  I'm a newbie on Functional Programming. But I'm changing that, trying hard at least. 2 things have been a great influence for me: Underscore.js and learning scala at coursera.org. Perhaps, at some level, I owe these new skills to Phil. RIP
drcode 14y ago
Phil's ideas on data structures are a critical corner stone in making functional programming easy and practical. Though most people may not have heard his name before, his influence on some computer languages was HUGE.
notlisted 14y ago
Might I inquire how/of what? I know it shouldn't matter, but I'm always curious when the cause of death is not listed. I've read too many of these where smart people decide to leave us of their own accord. WikiPedia was no help.
- bad_user 14y ago
  That's just the animal inside you that's hungry for gossip talking, because it really doesn't matter. And I know you'll come up with some kind of rationalization for wanting to know, but you should respect the wishes of family and close friends, whatever the reasons may be, especially because it really does not matter, the loss being the same with or without the reason your gossip-hungry brain may expect.
  - notlisted 14y ago
    Indeed, this has nothing to do with gossip, but everything with the prevalence of depression & burnout in the tech field. It was addressed by Joel from Fog creek several years ago.
sbochins 14y ago
I never heard of this guy, but after reading this I realized I know about the data structure he invited that is used in Clojure and other functional languages. We should celebrate the important impact he had on technology. His work will live on.
ffk 14y ago
Considering his contributions, can we have a black bar today?
millerm 14y ago
I had the pleasure of having a very long, beer filled conversation about tech, computer languages and culture with Phil while at ClojureConj in Raleigh last year. He was a very intelligent and charismatic man. It's amazing how sad the death of someone you had only known for a short moment can be.
pjscott 14y ago
Phil Bagwell's VList is a really cute data structure. It's an immutable list with O(1) cons and cdr, lookups that are expected constant time and O(lg n) worst-case time, and better cache locality than linked lists. The wikipedia entry is short and understandable, which makes it easy to honor the dead in a pleasant and enlightening way: http://en.wikipedia.org/wiki/VList http://en.wikipedia.org/wiki/VList Obviously a very clever guy.
tiarkrompf 14y ago
Here is a collection of all his papers: http://infoscience.epfl.ch/search?ln=en&p=bagwell http://infoscience.epfl.ch/search?ln=en&p=bagwell
joshhart 14y ago
I met Phil at Scaladays in 2011. He was such a nice guy. I remember having some ideas about how to improve the efficiency of hash array trie maps for certain cases and types of data and I remember how encouraging our conversation was. He'll be missed.

182 points by SanderMak 14y ago

See http://infoscience.epfl.ch/record/64410/files/techlists.pdf http://infoscience.epfl.ch/record/64410/files/techlists.pdf for one of Phil's most influential papers. Surprising (and saddening) to hear of his passing. See also http://lampwww.epfl.ch/~prokopec/ctries-snapshot.pdf http://lampwww.epfl.ch/~prokopec/ctries-snapshot.pdf for a more recent contribution.

marshray 14y ago
Wow, this Ctrie looks like a concurrent Swiss Army data structure. Very interesting, thanks.

fogus 14y ago

Some background on the kind of person Phil Bagwell was and his contributions http://blog.fogus.me/2012/10/15/phil-bagwell-rest-in-peace http://blog.fogus.me/2012/10/15/phil-bagwell-rest-in-peace Of particular note is his work creating Hash Array Mapped Tries (aka functional associative arrays) detailed in http://lampwww.epfl.ch/papers/idealhashtrees.pdf http://lampwww.epfl.ch/papers/idealhashtrees.pdf and implemented in Haskell, Clojure, Scala and Rubinius. His latest work involved lock-free versions called CTrees[1] that are likely to create more waves as others become familiar with them. [1]: https://github.com/axel22/Ctries https://github.com/axel22/Ctries

carterschonwald 14y ago
is there a good spot to find all the papers he's coauthored?
- SanderMak 14y ago
  Here's a start: http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/b/Bagwell:Phil.html http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/b/... doesn't include technical reports though.

noelwelsh 14y ago

I only met Phil once, but it was his suggestion that led to the formation of Underscore, where I now work. I hoped I'd get a chance to say thanks, but now that opportunity is gone.

berserkpi 14y ago
I'm a newbie on Functional Programming. But I'm changing that, trying hard at least. 2 things have been a great influence for me: Underscore.js and learning scala at coursera.org. Perhaps, at some level, I owe these new skills to Phil. RIP

drcode 14y ago

Phil's ideas on data structures are a critical corner stone in making functional programming easy and practical. Though most people may not have heard his name before, his influence on some computer languages was HUGE.

notlisted 14y ago

Might I inquire how/of what? I know it shouldn't matter, but I'm always curious when the cause of death is not listed. I've read too many of these where smart people decide to leave us of their own accord. WikiPedia was no help.

bad_user 14y ago
That's just the animal inside you that's hungry for gossip talking, because it really doesn't matter. And I know you'll come up with some kind of rationalization for wanting to know, but you should respect the wishes of family and close friends, whatever the reasons may be, especially because it really does not matter, the loss being the same with or without the reason your gossip-hungry brain may expect.
- notlisted 14y ago
  Indeed, this has nothing to do with gossip, but everything with the prevalence of depression & burnout in the tech field. It was addressed by Joel from Fog creek several years ago.

sbochins 14y ago

I never heard of this guy, but after reading this I realized I know about the data structure he invited that is used in Clojure and other functional languages. We should celebrate the important impact he had on technology. His work will live on.

ffk 14y ago

Considering his contributions, can we have a black bar today?

millerm 14y ago

I had the pleasure of having a very long, beer filled conversation about tech, computer languages and culture with Phil while at ClojureConj in Raleigh last year. He was a very intelligent and charismatic man. It's amazing how sad the death of someone you had only known for a short moment can be.

pjscott 14y ago

Phil Bagwell's VList is a really cute data structure. It's an immutable list with O(1) cons and cdr, lookups that are expected constant time and O(lg n) worst-case time, and better cache locality than linked lists. The wikipedia entry is short and understandable, which makes it easy to honor the dead in a pleasant and enlightening way: http://en.wikipedia.org/wiki/VList http://en.wikipedia.org/wiki/VList Obviously a very clever guy.

tiarkrompf 14y ago

Here is a collection of all his papers: http://infoscience.epfl.ch/search?ln=en&p=bagwell http://infoscience.epfl.ch/search?ln=en&p=bagwell

joshhart 14y ago

I met Phil at Scaladays in 2011. He was such a nice guy. I remember having some ideas about how to improve the efficiency of hash array trie maps for certain cases and types of data and I remember how encouraging our conversation was. He'll be missed.