Release Announcement for v6.8
96 views
Skip to first unread message
Vincent St-Amour
Jan 15, 2017, 2:14:55 PM1/15/17
to racke...@googlegroups.com
The release announcement sketch that I have so far is below. Please
mail me new items and/or edits.
Please phrase announcements using complete sentences and avoid the
word "now".
----------------------------------------------------------------------
Alexander Hedges:
- FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
Andrew Kent (filled in by Sam):
Typed Racket more efficiently represents and computes types and
propositions, resulting in significantly faster type checking for many
programs. For more details, see this blog post by Andrew Kent:
https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
Ben Greenman:
- scribble/html changes (1676671e @ scribble)
Gustavo Massaccesi
- optimizations (58333903 c3595c56 9ebfdb54 25dc89a2 @ racket)
Matthew Butterick:
- new logo
Matthew Flatt:
- touchbar mac support (b9e94f9c @ gui)
- support for GTK3 on wayland (48c1b282 @ gui)
- syntax property preservation (8f9d4860 @ racket)
- list operation space optimizations (5e94a906 d7b18e7a @ racket)
Ryan Culpepper:
- raco macro-stepper (ce7be92c @ macro-debugger)
William G Hatch:
- faster scrolling on GTK? (3b280551 @ gui)
----------------------------------------------------------------------
mail me new items and/or edits.
Please phrase announcements using complete sentences and avoid the
word "now".
----------------------------------------------------------------------
Alexander Hedges:
- FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
Andrew Kent (filled in by Sam):
Typed Racket more efficiently represents and computes types and
propositions, resulting in significantly faster type checking for many
programs. For more details, see this blog post by Andrew Kent:
https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
Ben Greenman:
- scribble/html changes (1676671e @ scribble)
Gustavo Massaccesi
- optimizations (58333903 c3595c56 9ebfdb54 25dc89a2 @ racket)
Matthew Butterick:
- new logo
Matthew Flatt:
- touchbar mac support (b9e94f9c @ gui)
- support for GTK3 on wayland (48c1b282 @ gui)
- syntax property preservation (8f9d4860 @ racket)
- list operation space optimizations (5e94a906 d7b18e7a @ racket)
Ryan Culpepper:
- raco macro-stepper (ce7be92c @ macro-debugger)
William G Hatch:
- faster scrolling on GTK? (3b280551 @ gui)
----------------------------------------------------------------------
Ben Greenman
Jan 15, 2017, 3:31:08 PM1/15/17
to Vincent St-Amour, Racket-Dev List
On Sun, Jan 15, 2017 at 2:14 PM, Vincent St-Amour <stam...@eecs.northwestern.edu> wrote:
Ben Greenman:
- scribble/html changes (1676671e @ scribble)
[[ I think this is too small to include, but here's a blurb ]]
The `scribble/html/extra` module provides renderers for HTML elements that are part of the WHATWG HTML standard [1] but not part by the `scribble/html` language by default (e.g., `map`, `time`, `video`).
Gustavo Massaccesi
Jan 15, 2017, 3:51:07 PM1/15/17
to Vincent St-Amour, Ben Greenman, Racket-Dev List
- c3595c56 9ebfdb54 25dc89a2
I think they are too minor.
- 58333903
I like it, specially because it's the result of many intermediate
steps, like 7c1cb1a2. But it's difficult to explain (specially because
it depends on the knowledge that equal? is much slower than eq?). So I
think it's better to ignore it in the announcement.
I propose to add:
- f159295e
*** The bytecode compiler detects more optimization opportunities,
including when variables always hold a Boolean value. In particular in
expressions like (or (symbol? x) (something)), the optimizations
avoids the creation of an intermediate variable to store the result of
(symbol? x).
- 9c1b8707(mflat) and 6d1018fb
*** The bytecode optimizer avoids dropping the stack frames in some
reductions. This will provide better stack traces for errors when
errotrace is enable.
Gustavo
I think they are too minor.
- 58333903
I like it, specially because it's the result of many intermediate
steps, like 7c1cb1a2. But it's difficult to explain (specially because
it depends on the knowledge that equal? is much slower than eq?). So I
think it's better to ignore it in the announcement.
I propose to add:
- f159295e
*** The bytecode compiler detects more optimization opportunities,
including when variables always hold a Boolean value. In particular in
expressions like (or (symbol? x) (something)), the optimizations
avoids the creation of an intermediate variable to store the result of
(symbol? x).
- 9c1b8707(mflat) and 6d1018fb
*** The bytecode optimizer avoids dropping the stack frames in some
reductions. This will provide better stack traces for errors when
errotrace is enable.
Gustavo
Vincent St-Amour
Jan 18, 2017, 1:08:58 PM1/18/17
to Gustavo Massaccesi, Vincent St-Amour, Ben Greenman, Racket-Dev List
On Sun, 15 Jan 2017 14:50:25 -0600,
Gustavo Massaccesi wrote:
> - 58333903
> I like it, specially because it's the result of many intermediate
> steps, like 7c1cb1a2. But it's difficult to explain (specially because
> it depends on the knowledge that equal? is much slower than eq?). So I
> think it's better to ignore it in the announcement.
I think it's safe to assume (or very briefly state) that `equal?` can be
Gustavo Massaccesi wrote:
> - 58333903
> I like it, specially because it's the result of many intermediate
> steps, like 7c1cb1a2. But it's difficult to explain (specially because
> it depends on the knowledge that equal? is much slower than eq?). So I
> think it's better to ignore it in the announcement.
much more expensive than `eq?`. But we can leave it out if you prefer.
Vincent
Gustavo Massaccesi
Jan 19, 2017, 1:29:40 PM1/19/17
to Vincent St-Amour, Ben Greenman, Racket-Dev List
What about something like:
The bytecode optimizer can reduce more expressions like (equal? x y)
to (eqv? x y) or (eq? x y), in case the type of x or y is known and
the comparison is equivalent for that type.
Gustavo
The bytecode optimizer can reduce more expressions like (equal? x y)
to (eqv? x y) or (eq? x y), in case the type of x or y is known and
the comparison is equivalent for that type.
Gustavo
Vincent St-Amour
Jan 19, 2017, 4:02:50 PM1/19/17
to Gustavo Massaccesi, Vincent St-Amour, Ben Greenman, Racket-Dev List
Sounds good! Thanks.
Vincent
Vincent
William Hatch
Jan 20, 2017, 7:04:03 AM1/20/17
to racke...@googlegroups.com
The gtk scrolling fix is probably not worth noting on the announcement. And actually you might call it slower scrolling.
On January 15, 2017 12:14:55 PM MST, Vincent St-Amour <stam...@eecs.northwestern.edu> wrote:
The release announcement sketch that I have so far is below. Please
mail me new items and/or edits.
Please phrase announcements using complete sentences and avoid the
word "now".
Alexander Hedges:
- FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
Andrew Kent (filled in by Sam):
Typed Racket more efficiently represents and computes types and
propositions, resulting in significantly faster type checking for many
programs. For more details, see this blog post by Andrew Kent:
https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
Ben Greenman:
- scribble/html changes (1676671e @ scribble)
Gustavo Massaccesi
- optimizations (58333903 c3595c56 9ebfdb54 25dc89a2 @ racket)
Matthew Butterick:
- new logo
Matthew Flatt:
- touchbar mac support (b9e94f9c @ gui)
- support for GTK3 on wayland (48c1b282 @ gui)
- syntax property preservation (8f9d4860 @ racket)
- list operation space optimizations (5e94a906 d7b18e7a @ racket)
Ryan Culpepper:
- raco macro-stepper (ce7be92c @ macro-debugger)
William G Hatch:
- faster scrolling on GTK? (3b280551 @ gui)
--
Sent from my pocket computer. Please excuse the lint.
Vincent St-Amour
Jan 20, 2017, 4:43:50 PM1/20/17
to racke...@googlegroups.com
Here's what I have so far.
- Typed Racket more efficiently represents and computes types and
`(equal? x y)` to `(eqv? x y)` or `(eq? x y)` when the type of `x` or
`y` is known.
- The bytecode optimizer avoids dropping stack frames in some reductions.
This provides better stack traces for errors when errotrace is enabled.
- The bytecode compiler detects more optimization opportunities,
including when variables always hold a boolean value. In particular in
that are part of the WHATWG HTML standard but not part of the `scribble/html`
`match` form.
- In the DeinProgramm languages, define-record-procedures supports
specifying signatures as part of the form, instantly enabling
properties quantifying over such records.
If you haven't filled in your items, please do so.
Alexander Hedges:
- FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
> --
> You received this message because you are subscribed to the Google Groups "Racket Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-dev+...@googlegroups.com.
> To post to this group, send email to racke...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-dev/m2fukk3zkg.wl-stamourv%40eecs.northwestern.edu.
> For more options, visit https://groups.google.com/d/optout.
- Typed Racket more efficiently represents and computes types and
propositions, resulting in significantly faster type checking for many
programs. For more details, see this blog post by Andrew Kent:
https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
- The bytecode optimizer can reduce more expressions of the form
programs. For more details, see this blog post by Andrew Kent:
https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
`(equal? x y)` to `(eqv? x y)` or `(eq? x y)` when the type of `x` or
`y` is known.
- The bytecode optimizer avoids dropping stack frames in some reductions.
This provides better stack traces for errors when errotrace is enabled.
- The bytecode compiler detects more optimization opportunities,
including when variables always hold a boolean value. In particular in
expressions like (or (symbol? x) (something)), the optimizations
avoids the creation of an intermediate variable to store the result of
(symbol? x).
- The `scribble/html/extra` module provides renderers for HTML elements
avoids the creation of an intermediate variable to store the result of
(symbol? x).
that are part of the WHATWG HTML standard but not part of the `scribble/html`
`scribble/html` language by default (e.g., `map`, `time`, `video`).
- The DeinProgramm teaching languages support pattern matching using the
`match` form.
- In the DeinProgramm languages, define-record-procedures supports
specifying signatures as part of the form, instantly enabling
properties quantifying over such records.
If you haven't filled in your items, please do so.
Alexander Hedges:
- FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
Matthew Butterick:
- new logo
Matthew Flatt:
- touchbar mac support (b9e94f9c @ gui)
- support for GTK3 on wayland (48c1b282 @ gui)
- syntax property preservation (8f9d4860 @ racket)
- list operation space optimizations (5e94a906 d7b18e7a @ racket)
Ryan Culpepper:
- raco macro-stepper (ce7be92c @ macro-debugger)
Vincent
- new logo
Matthew Flatt:
- touchbar mac support (b9e94f9c @ gui)
- support for GTK3 on wayland (48c1b282 @ gui)
- syntax property preservation (8f9d4860 @ racket)
- list operation space optimizations (5e94a906 d7b18e7a @ racket)
Ryan Culpepper:
- raco macro-stepper (ce7be92c @ macro-debugger)
> You received this message because you are subscribed to the Google Groups "Racket Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-dev+...@googlegroups.com.
> To post to this group, send email to racke...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-dev/m2fukk3zkg.wl-stamourv%40eecs.northwestern.edu.
> For more options, visit https://groups.google.com/d/optout.
Leif Andersen
Jan 20, 2017, 4:47:49 PM1/20/17
to William Hatch, racket-dev
Wait, slower scrolling?
Or do you mean more responsive but slower scrolling?--
You received this message because you are subscribed to the Google Groups "Racket Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to racket-dev+unsubscribe@googlegroups.com.
To post to this group, send email to racke...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/racket-dev/71BAAA55-FE88-432B-992C-0A4CE7204909%40hatch.uno.
Andrew Kent
Jan 20, 2017, 4:51:39 PM1/20/17
to Racket Developers, stam...@eecs.northwestern.edu
I think dropping "significantly" from the Typed Racket blurb is probably more accurate.
William G Hatch
Jan 20, 2017, 5:02:22 PM1/20/17
to Leif Andersen, racket-dev
On Fri, Jan 20, 2017 at 04:47:07PM -0500, Leif Andersen wrote:
>Wait, slower scrolling?
>
>Or do you mean more responsive but slower scrolling?
That's probably a better way of saying it. Some devices give "smooth
>Wait, slower scrolling?
>
>Or do you mean more responsive but slower scrolling?
scroll" events with a float amount, eg. -0.15X and +0.05Y. But Racket
doesn't support smooth scrolling and each of these were translated into
a full scroll event, meaning scrolling was in the ballpark of 4-10 times
too fast and was jumpy (even trying to move very slowly on my trackpad
would send multiple scroll events). I simply made it accumulate the
scroll events until they got to +/-1.0 before giving a Racket scroll
event, so it scrolls more consistently with everything else.
>
>Or am I missing something here?
>
>
>~Leif Andersen
>
>On Wed, Jan 18, 2017 at 1:51 AM, William Hatch <wil...@hatch.uno> wrote:
>
>> The gtk scrolling fix is probably not worth noting on the announcement.
>> And actually you might call it slower scrolling.
>>
>> On January 15, 2017 12:14:55 PM MST, Vincent St-Amour <
>> stam...@eecs.northwestern.edu> wrote:
>>>
>>> The release announcement sketch that I have so far is below. Please
>>> mail me new items and/or edits.
>>>
>>> Please phrase announcements using complete sentences and avoid the
>>> word "now".
>>>
>>>
>>> Alexander Hedges:
>>> - FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
>>>
>>> Andrew Kent (filled in by Sam):
>>> Typed Racket more efficiently represents and computes types and
>>> propositions, resulting in significantly faster type checking for many
>>> programs. For more details, see this blog post by Andrew Kent:
>>> https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
>>>
>>> Ben Greenman:
>>> - scribble/html changes (1676671e @ scribble)
>>>
>>> Gustavo Massaccesi
>>> - optimizations (58333903 c3595c56 9ebfdb54 25dc89a2 @ racket)
>>>
>>> Matthew Butterick:
>>> - new logo
>>>
>>> Matthew Flatt:
>>> - touchbar mac support (b9e94f9c @ gui)
>>> - support for GTK3 on wayland (48c1b282 @ gui)
>>> - syntax property preservation (8f9d4860 @ racket)
>>> - list operation space optimizations (5e94a906 d7b18e7a @ racket)
>>>
>>> Ryan Culpepper:
>>> - raco macro-stepper (ce7be92c @ macro-debugger)
>>>
>>> William G Hatch:
>>> - faster scrolling on GTK? (3b280551 @ gui)
>>>
>>> ------------------------------
>>>
>>> Alexander Hedges:
>>> - FreeBSD PowerPC support (d031f30d f1d853ee @ racket)
>>>
>>> Andrew Kent (filled in by Sam):
>>> Typed Racket more efficiently represents and computes types and
>>> propositions, resulting in significantly faster type checking for many
>>> programs. For more details, see this blog post by Andrew Kent:
>>> https://pnwamk.github.io/blog/2017/01/08/some-changes-in-typed-racket-v6-8/
>>>
>>> Ben Greenman:
>>> - scribble/html changes (1676671e @ scribble)
>>>
>>> Gustavo Massaccesi
>>> - optimizations (58333903 c3595c56 9ebfdb54 25dc89a2 @ racket)
>>>
>>> Matthew Butterick:
>>> - new logo
>>>
>>> Matthew Flatt:
>>> - touchbar mac support (b9e94f9c @ gui)
>>> - support for GTK3 on wayland (48c1b282 @ gui)
>>> - syntax property preservation (8f9d4860 @ racket)
>>> - list operation space optimizations (5e94a906 d7b18e7a @ racket)
>>>
>>> Ryan Culpepper:
>>> - raco macro-stepper (ce7be92c @ macro-debugger)
>>>
>>> William G Hatch:
>>> - faster scrolling on GTK? (3b280551 @ gui)
>>>
>>>
>>>
>> --
>> Sent from my pocket computer. Please excuse the lint.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Racket Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to racket-dev+...@googlegroups.com.
>>>
>> --
>> Sent from my pocket computer. Please excuse the lint.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Racket Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> To post to this group, send email to racke...@googlegroups.com.
>> To view this discussion on the web visit https://groups.google.com/d/
>> msgid/racket-dev/71BAAA55-FE88-432B-992C-0A4CE7204909%40hatch.uno
>> <https://groups.google.com/d/msgid/racket-dev/71BAAA55-FE88-432B-992C-0A4CE7204909%40hatch.uno?utm_medium=email&utm_source=footer>
>> To view this discussion on the web visit https://groups.google.com/d/
>> msgid/racket-dev/71BAAA55-FE88-432B-992C-0A4CE7204909%40hatch.uno
>> .
Reply all
Reply to author
Forward
0 new messages