Isn't a stack O(1) time for push & pop and O(n) space? What is the tradeoff being made here, where both space and time complexity increased? Or, am I missing something about the meaning of "n + O(log)", where the 'n' is not inside an 'O'. I don't really know what "n + O(log)" means exactly. On Wed, Sep 2, 2020 at 12:21 PM David Feuer wrote:

Typical stacks, queues, and deques are designed to have O(1) operations. These ones experiment in a different direction.

On Wed, Sep 2, 2020, 12:19 AM ☂Josh Chia (謝任中) wrote:

...

Could you elaborate on what the package does? The description is "Stacks, queues, and deques that take n + O(log n) space at the cost of having amortized O(log n) time complexity for basic operations." But why is it a 'cost' to have O(log n) instead of O(n) cost for basic operations on list-like structures (such as insert and delete presumably)? So, I didn't understand what the package is for.

Do you mean "take amortized O(log n) time for basic operations at the cost of n + O(log n) space"? Or is it something else I didn't imagine?

On Wed, Sep 2, 2020 at 4:17 AM David Feuer wrote:

...

I am pleased to release the second version of the compact-sequences package, now with deques!

Changes: * Add deques. * Change operator precedence. * Add a test suite. Thanks to David Himmelstrup for setting up the test and CI framework. * Clean up internals somewhat. * Add a proof of amortized bounds for the stack implementation. Thanks, Li-Yao Xia.

There are still plenty of things to work on, and help is always welcome.

David Feuer _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.

That explanation makes sense. It's probably worth adding a couple of sentences to that effect to the package description. I haven't seen that notation before, so I wasn't sure what it meant before you explained it here. On Tue, Sep 1, 2020 at 9:57 PM David Feuer wrote:

Yes, the n not being inside the O is the point. These structures hold n pointers in n words, plus a logarithmic term. Something like a linked list or Data.Sequence will hold n pointers in around c*n words, where c>1. For a list, c=3.

On Wed, Sep 2, 2020, 12:26 AM ☂Josh Chia (謝任中) wrote:

...

Isn't a stack O(1) time for push & pop and O(n) space? What is the tradeoff being made here, where both space and time complexity increased?

Or, am I missing something about the meaning of "n + O(log)", where the 'n' is not inside an 'O'. I don't really know what "n + O(log)" means exactly.

On Wed, Sep 2, 2020 at 12:21 PM David Feuer wrote:

...

Typical stacks, queues, and deques are designed to have O(1) operations. These ones experiment in a different direction.

On Wed, Sep 2, 2020, 12:19 AM ☂Josh Chia (謝任中) wrote:

...

Could you elaborate on what the package does? The description is "Stacks, queues, and deques that take n + O(log n) space at the cost of having amortized O(log n) time complexity for basic operations." But why is it a 'cost' to have O(log n) instead of O(n) cost for basic operations on list-like structures (such as insert and delete presumably)? So, I didn't understand what the package is for.

Do you mean "take amortized O(log n) time for basic operations at the cost of n + O(log n) space"? Or is it something else I didn't imagine?

On Wed, Sep 2, 2020 at 4:17 AM David Feuer wrote:

...

I am pleased to release the second version of the compact-sequences package, now with deques!

Changes: * Add deques. * Change operator precedence. * Add a test suite. Thanks to David Himmelstrup for setting up the test and CI framework. * Clean up internals somewhat. * Add a proof of amortized bounds for the stack implementation. Thanks, Li-Yao Xia.

There are still plenty of things to work on, and help is always welcome.

David Feuer _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.

_______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.