Giuseppe Ottaviano

I work with computers; I do a lot of reading, looking at graphs, and a bit of typing.

[Giuseppe Ottaviano @ot_y]

@flaminhottweets @jbarro Evergreen post https://t.co/y9qZtdVgtL

[Giuseppe Ottaviano @ot_y]

@fortnow Why do we still call it paper if it's not made out of papyrus

[Giuseppe Ottaviano @ot_y]

@ZachWeiner https://t.co/BavOMRtCZ0

[Giuseppe Ottaviano @ot_y]

@ericniebler My gut feeling is that I agree with your intention, as cannot think of any weird type that does that, but formally defining what "specified moved-from state" means might be tricky and conflict with legitimate use cases.

[Spinoza @spinozait]

Secondo me i nuovi tappi non piacciono alla destra perché quando li sviti restano appesi a testa in giù.
[@chiaravix]

[Giuseppe Ottaviano @ot_y]

@triketora Yes.

[Giuseppe Ottaviano @ot_y]

@RiccardoTrezzi Se normalizzato su GDP la situazione sembra meno drammatica?

[Giuseppe Ottaviano @ot_y]

@freezydorito There’s a whole ecosystem and civilization in there by now.

[Giuseppe Ottaviano @ot_y]

@gentile_giulia As opposed to law?

[Giuseppe Ottaviano @ot_y]

@gentile_giulia Every other email I get starts with “Hi Guiseppe”

[Giuseppe Ottaviano @ot_y]

@d_feldman They should open-source it and call it https://t.co/P63U4VOriY

[Giuseppe Ottaviano @ot_y]

@triketora It is a concern people have

https://t.co/IlTEIQcdQZ

[Giuseppe Ottaviano @ot_y]

@MarkCallaghanDB If you don't cite the prior art in other areas you're silently implying you're the first one to come up with the idea, instead of sharing credit.

[Giuseppe Ottaviano @ot_y]

@MarkCallaghanDB My comment was just pointing to prior art, hence the "for example", I didn't say "use folly".
And novelty is definitely a factor when evaluating a paper, especially if it doesn't have a thorough review of related work. But this was a workshop paper, so the bar is likely lower.

[Giuseppe Ottaviano @ot_y]

@triketora The empiricists get stuck on trivial problems (or introduce trivial bugs) that could be solved by spending 5 minutes reading the definition.
The formalists get stuck in endless nitpicking about the correct definitions.
The former are usually more productive.

[Giuseppe Ottaviano @ot_y]

@triketora Oversimplifying things a lot, I think that there are two camps in programming, the empiricists (try and see what happens) and the formalists (strictly derive from the definitions). Each camp finds the way the other operates incomprehensible and infuriating.

[Giuseppe Ottaviano @ot_y]

@MarkCallaghanDB Also the idea of using a pool of memory that is never returned to the OS so it can be used by concurrent algorithms that use version numbers to detect validity is not new, see for example IndexedMemPool in folly and its uses.

[Giuseppe Ottaviano @ot_y]

@MarkCallaghanDB The analysis and experiments are based on a flawed understanding of hazard pointers, readers don’t need sequentially consistent writes, a compiler barrier is enough (with a membarrier on reclamation), so their overhead is off by an order of magnitude.

[Giuseppe Ottaviano @ot_y]

@Jeremy_thegoose @lachlanjhodson @chaser The iceberg was too.