Message from discussion
Cutting 1.31
Received: by 10.14.122.69 with SMTP id s45mr39405eeh.7.1329347201148;
Wed, 15 Feb 2012 15:06:41 -0800 (PST)
X-BeenThere: vimium-dev@googlegroups.com
Received: by 10.14.52.76 with SMTP id d52ls2455eec.6.canary; Wed, 15 Feb 2012
15:06:40 -0800 (PST)
Received: by 10.14.17.222 with SMTP id j70mr45173eej.2.1329347200701;
Wed, 15 Feb 2012 15:06:40 -0800 (PST)
Received: by 10.14.17.222 with SMTP id j70mr45172eej.2.1329347200659;
Wed, 15 Feb 2012 15:06:40 -0800 (PST)
Return-Path: <phil.cro...@gmail.com>
Received: from mail-ee0-f54.google.com (mail-ee0-f54.google.com [74.125.83.54])
by gmr-mx.google.com with ESMTPS id n48si3369995eeh.1.2012.02.15.15.06.40
(version=TLSv1/SSLv3 cipher=OTHER);
Wed, 15 Feb 2012 15:06:40 -0800 (PST)
Received-SPF: pass (google.com: domain of phil.cro...@gmail.com designates 74.125.83.54 as permitted sender) client-ip=74.125.83.54;
Authentication-Results: gmr-mx.google.com; spf=pass (google.com: domain of phil.cro...@gmail.com designates 74.125.83.54 as permitted sender) smtp.mail=phil.cro...@gmail.com; dkim=pass header...@gmail.com
Received: by eekb47 with SMTP id b47so628773eek.13
for <vimium-dev@googlegroups.com>; Wed, 15 Feb 2012 15:06:40 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=gmail.com; s=gamma;
h=mime-version:in-reply-to:references:date:message-id:subject:from:to
:content-type;
bh=JaP7Pml+ftW95X+h9hG8cCkxBq61subYzGmYn9q5Wj4=;
b=NlhTKfVU+n/wJoRujPvgOMrcekvKvrZUz94FicXufDiwJOXIWqSl8jbJWOhGZbioM7
NAa2FOWvyHCF+J4eeVIMbtBjqiIcBb8MSxgMhqW7bGffnTR7HfOLEdoLsyA1dL9s3TNa
czWAcX/HvKOIqVJr5pOPELMtASjKXgEK8PewY=
MIME-Version: 1.0
Received: by 10.14.95.8 with SMTP id o8mr22994eef.111.1329347200484; Wed, 15
Feb 2012 15:06:40 -0800 (PST)
Received: by 10.213.31.212 with HTTP; Wed, 15 Feb 2012 15:06:40 -0800 (PST)
In-Reply-To: <CABVpgLWhi009YO6cZ3RyYf2hXrtTRkucXOopJdO_d7K_s5PGNw@mail.gmail.com>
References: <CABVpgLXLws1C1k5KPb=H3RY3-LjMwmz4EZNZJ_+kACCeoY6cQQ@mail.gmail.com>
<CAADJaVUPvdLLrRe_1SidqfhPPbAhTFEYAa+Zdy+SzyJvmMo...@mail.gmail.com>
<CA+V-=dp3TJcWR+uKo+SznKN26zw=FkGnDmU6QpJ2v8Km3-E...@mail.gmail.com>
<CAADJaVWT5nBJXrYB51e30UC0FJ329jOOuChGCX9MgX98Q0t...@mail.gmail.com>
<CABVpgLWhi009YO6cZ3RyYf2hXrtTRkucXOopJdO_d7K_s5P...@mail.gmail.com>
Date: Wed, 15 Feb 2012 15:06:40 -0800
Message-ID: <CAADJaVUns+Mo4cC_yGx406ttcQZoMYO4+pjZnROny5NiR+Y...@mail.gmail.com>
Subject: Re: Cutting 1.31
From: Phil Crosby <phil.cro...@gmail.com>
To: vimium-dev@googlegroups.com
Content-Type: multipart/alternative; boundary=bcaec52be479ab051404b908c3c6
--bcaec52be479ab051404b908c3c6
Content-Type: text/plain; charset=ISO-8859-1
I merged in a pull request to hide the HUD (which obscures Facebook Chat's
textbox) until we come up with a longer term solution, like eliminating the
HUD. The 1.31 milestone now has no open issues.
For #3, let's make the docs as lightweight as possible. I didn't expect
that they'd be changed that much which is why I suggested simply a link to
the wiki.
Let me know when we're clear from your end Jez and we'll push this out.
On Mon, Feb 13, 2012 at 2:25 PM, Jez <jezr...@gmail.com> wrote:
> #1: Alright, I'll reverse it in my next push.
>
> #2: Oops, I linked to the wrong issue. It was supposed to be #426.
> Essentially, I made a commit that returned focus to the top-level frame if
> ESC was pressed more than once. However, this turns out to be the wrong
> thing to do at times -- sometimes the top-level frame is not the one with
> the scrollable content. I have a bunch of ideas to deal with this, but I
> don't have time to experiment with them at the moment.
>
> #3: I was thinking of linking to a page that's packaged within the
> extension, actually. Linking to our wiki might cause confusion as the dev
> version deviates from the last released one.
>
> #4: I've replied in the comments on #390<https://github.com/philc/vimium/issues/390>
> .
>
> Thanks so much for working through all those issues last night! I'll
> follow up on them this coming weekend.
>
> Cheers,
> Jez
>
>
>
> On Mon, Feb 13, 2012 at 2:21 AM, Phil Crosby <phil.cro...@gmail.com>wrote:
>
>> #2: I closed #138 based on the discussion. I didn't see any mention of a
>> patch or next actions.
>>
>> #3: Nice idea. We can replace the "found a bug? report it here" link with
>> "Vimium documentation" which links to our wiki.
>>
>> #4: note that Chrome has the shortcut CMD+1 for the first tab. That means
>> g+x and CMD+x will be offset by 1 and thus frustrating to use together.
>> I've added a note about this on #390 (
>> https://github.com/philc/vimium/issues/390).
>>
>> I've scrubbed through the issues in the issue queue and we're looking
>> good. I've moved two issues into a milestone that we should close these
>> down before cutting the 1.31 release:
>>
>> https://github.com/philc/vimium/issues?sort=updated&direction=desc&state=open&page=1&milestone=1
>>
>> On Sun, Feb 12, 2012 at 10:07 PM, Ilya Sukhar <ilya.suk...@gmail.com>wrote:
>>
>>> Thanks so much for all your work Jez. Regarding your points:
>>>
>>> #1: I commented on the bug but unfortunately I have nothing productive
>>> to add. I'm okay with reversing this.
>>> #4: I'm fine with numbering but g0 for the last tab doesn't seem right.
>>> g0 should be the first tab.
>>>
>>> Ilya
>>>
>>>
>>> On Sat, Feb 11, 2012 at 5:22 PM, Phil Crosby <phil.cro...@gmail.com>wrote:
>>>
>>>> Thanks for taking the lead on this Jez. The linkhint improvements
>>>> really improve quality of life (thanks for everyone who contributed), so we
>>>> shouldn't keep them out of users' hands for too long.
>>>>
>>>> Regarding this change, I thought you concluded after experimenting that
>>>> it wasn't gelling that welll:
>>>> Insert mode is automatically exited when changing tabs
>>>>
>>>> It would be good to tag issues with the 1.31 milestone for all of the
>>>> remaining bugs we should resolve prior to cutting a new release. I can
>>>> scrub through the queue later tonight, apply the tag, and see if there's
>>>> anything big we're missing.
>>>>
>>>>
>>>> On Sat, Feb 11, 2012 at 4:27 PM, Jez <jezr...@gmail.com> wrote:
>>>>
>>>>> There have been quite a lot of good changes lately and I think it
>>>>> would be great if we could get that pushed out to our users.
>>>>>
>>>>> Here's a summary of the changelog:
>>>>>
>>>>> - Hint appearance has undergone a redesign
>>>>> - New hint scattering strategy for alphabetic hints
>>>>> - Divs can now be scrolled after the user clicks on them (we'll
>>>>> eventually add a pure keyboard mechanism for doing this)
>>>>> - Detection and following of next / prev links has been improved
>>>>> - Addition of g0 and g$ commands
>>>>> - Addition of p/P for URL pasting
>>>>> - 'Hardening' of CSS to prevent site interference
>>>>> - A greatly enhanced find mode, in particular:
>>>>> - Regex support (turned on by the \r escape sequence)
>>>>> - Smartcase searching is the default now (case sensitivity is
>>>>> triggered by uppercase characters)
>>>>> - Case sensitivity can be enforced using the \I escape sequence
>>>>> - Find queries are now global across tabs and persistent across
>>>>> browser sessions
>>>>> - Page shortcuts e.g. on GitHub should no longer be triggered
>>>>> during find mode
>>>>> - Find mode no longer triggers insert mode if there are input
>>>>> elements that match the query, but we can enter insert mode on one of the
>>>>> results by hitting <esc>.
>>>>> - Insert mode is automatically exited when changing tabs
>>>>> - Many many many bugfixes and minor improvements.
>>>>>
>>>>> Things I'd like to see settled before the push:
>>>>>
>>>>> - Escape in insert mode should not be passed to page --
>>>>> https://github.com/philc/vimium/issues/468 (Waiting for Ilya's
>>>>> input on this -- If you don't remember the reason for the original commit,
>>>>> could we just back it out?)
>>>>> - Have escape focus content --
>>>>> https://github.com/philc/vimium/issues/138 -- current
>>>>> implementation / interface is awkward, I plan to back it out before the
>>>>> release
>>>>> - Find mode behavior needs to be documented -- I'm thinking of
>>>>> linking to a separate HTML page from our help dialog
>>>>> - I don't like `g0` and `g$`. g$ in particular seems *really*
>>>>> awkward to press. Could we do simple tab numbering instead? I.e. g1 for the
>>>>> first tab, g2 for the second,, and g0 for the last tab. We can also add in
>>>>> negative numbers -- `g-1` for the second tab from the end, etc. It's pretty
>>>>> similar to Chrome / FF's default Ctrl/Cmd-[number] behavior for switching
>>>>> tabs, but a) we'll have negative numbers and b) `g` is considerably less
>>>>> awkward than `Ctrl`, which would benefit our non-Mac users. Or we could
>>>>> back this out as well and do some experimentation before our next release.
>>>>>
>>>>> Jez
>>>>>
>>>>
>>>>
>>>
>>
>
--bcaec52be479ab051404b908c3c6
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
I merged in a pull request to hide the HUD (which obscures Facebook Chat=
9;s textbox) until we come up with a longer term solution, like eliminating=
the HUD. The 1.31 milestone now has no open issues.<div><br></div><div>
For #3, let's make the docs as lightweight as possible. I didn't ex=
pect that they'd be changed that much which is why I suggested simply a=
link to the wiki.<br><div><br></div><div>Let me know when we're clear =
from your end Jez and we'll push this out.<br>
<br><div class=3D"gmail_quote">On Mon, Feb 13, 2012 at 2:25 PM, Jez <span d=
ir=3D"ltr"><<a href=3D"mailto:jezr...@gmail.com">jezr...@gmail.com</a>&g=
t;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0=
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>#1: Alright, I'll reverse it in my next push.</div><div><br></div>=
#2: Oops, I linked to the wrong issue. It was supposed to be #426. Essentia=
lly, I made a commit that returned focus to the top-level frame if ESC was =
pressed more than once. However, this turns out to be the wrong thing to do=
at times -- sometimes the top-level frame is not the one with the scrollab=
le content. I have a bunch of ideas to deal with this, but I don't have=
time to experiment with them at the moment.<div>
<br></div><div>#3: I was thinking of linking to a page that's packaged =
within the extension, actually. Linking to our wiki might cause confusion a=
s the dev version deviates from the last released one.</div><div><br></div>
<div>#4: I've replied in the comments on=A0<a href=3D"https://github.co=
m/philc/vimium/issues/390" target=3D"_blank">#390</a>.</div><div><br></div>=
<div>Thanks so much for working through all those issues last night! I'=
ll follow up on them this coming weekend.</div>
<div><br></div><div>Cheers,<br clear=3D"all">Jez<div><div class=3D"h5"><br>
<br><br><div class=3D"gmail_quote">On Mon, Feb 13, 2012 at 2:21 AM, Phil Cr=
osby <span dir=3D"ltr"><<a href=3D"mailto:phil.cro...@gmail.com" target=
=3D"_blank">phil.cro...@gmail.com</a>></span> wrote:<br><blockquote clas=
s=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;pad=
ding-left:1ex">
<div>#2: I closed #138 based on the discussion. I didn't see any mentio=
n of a patch or next actions.</div><div><br></div><div>#3: Nice idea. We ca=
n replace the "found a bug? report it here" link with "Vimiu=
m documentation" which links to our wiki.</div>
<div><br></div>#4: note that Chrome has the shortcut CMD+1 for the first ta=
b. That means g+x and CMD+x will be offset by 1 and thus frustrating to use=
together. I've added a note about this on #390 (<a href=3D"https://git=
hub.com/philc/vimium/issues/390" target=3D"_blank">https://github.com/philc=
/vimium/issues/390</a>).<div>
<br></div><div>I've scrubbed through the issues in the issue queue and =
we're looking good. I've moved two issues into a milestone that we =
should close these down before cutting the 1.31 release:</div><div><a href=
=3D"https://github.com/philc/vimium/issues?sort=3Dupdated&direction=3Dd=
esc&state=3Dopen&page=3D1&milestone=3D1" target=3D"_blank">http=
s://github.com/philc/vimium/issues?sort=3Dupdated&direction=3Ddesc&=
state=3Dopen&page=3D1&milestone=3D1</a></div>
<div><div>
<div><div><br><div class=3D"gmail_quote">On Sun, Feb 12, 2012 at 10:07 PM, =
Ilya Sukhar <span dir=3D"ltr"><<a href=3D"mailto:ilya.suk...@gmail.com" =
target=3D"_blank">ilya.suk...@gmail.com</a>></span> wrote:<br><blockquot=
e class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc sol=
id;padding-left:1ex">
Thanks so much for all your work Jez. Regarding your points:<div><br></div>=
<div>#1: I commented on the bug but unfortunately I have nothing productive=
to add. I'm okay with reversing this.</div><div>#4: I'm fine with =
numbering but g0 for the last tab doesn't seem right. g0 should be the =
first tab.</div>
<span><font color=3D"#888888">
<div><br></div></font></span><div><span><font color=3D"#888888">Ilya</font>=
</span><div><div><br><div><div><br><div class=3D"gmail_quote">On Sat, Feb 1=
1, 2012 at 5:22 PM, Phil Crosby <span dir=3D"ltr"><<a href=3D"mailto:phi=
l.cro...@gmail.com" target=3D"_blank">phil.cro...@gmail.com</a>></span> =
wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
Thanks for taking the lead on this Jez. The linkhint improvements really im=
prove quality of life (thanks for everyone who contributed), so we shouldn&=
#39;t keep them out of users' hands for too long.<div><br></div><div>
Regarding this change, I thought you concluded after experimenting that it =
wasn't gelling that welll:</div><div><div>Insert mode is automatically =
exited when changing tabs</div><div><br></div></div><div>It would be good t=
o tag issues with the 1.31 milestone for all of the remaining bugs we shoul=
d resolve prior to cutting a new release. I can scrub through the queue lat=
er tonight, apply the tag, and see if there's anything big we're mi=
ssing.<div>
<div></div><div><br>
<br><div class=3D"gmail_quote">On Sat, Feb 11, 2012 at 4:27 PM, Jez <span d=
ir=3D"ltr"><<a href=3D"mailto:jezr...@gmail.com" target=3D"_blank">jezre=
e...@gmail.com</a>></span> wrote:<br><blockquote class=3D"gmail_quote" sty=
le=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, san=
s-serif"><span style=3D"line-height:20px"><div>There have been quite a lot =
of good changes lately and I think it would be great if we could get that p=
ushed out to our users.</div>
<div><br></div><div>Here's a summary of the changelog:</div></span></fo=
nt></div><ul style=3D"margin-top:15px;margin-right:0px;margin-bottom:15px;m=
argin-left:0px;padding-top:0px;padding-right:0px;padding-bottom:0px;padding=
-left:30px;background-color:rgb(251,251,251)">
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Hint appearance has undergone a re=
design</span></font></li><li><font color=3D"#333333" face=3D"helvetica, ari=
al, freesans, clean, sans-serif"><span style=3D"line-height:20px">New hint =
scattering strategy for alphabetic hints</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Divs can now be scrolled after the=
user clicks on them (we'll eventually add a pure keyboard mechanism fo=
r doing this)</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Detection and following of next / =
prev links has been improved</span></font></li><li><font color=3D"#333333" =
face=3D"helvetica, arial, freesans, clean, sans-serif"><span style=3D"line-=
height:20px">Addition of g0 and g$ commands</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Addition of p/P for URL pasting</s=
pan></font></li><li><font color=3D"#333333" face=3D"helvetica, arial, frees=
ans, clean, sans-serif"><span style=3D"line-height:20px">'Hardening'=
; of CSS to prevent site interference</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px"><span style=3D"font-size:13px">A g=
reatly enhanced find mode, in particular:</span></span></font></li><ul><li>=
<span style=3D"color:rgb(51,51,51);font-family:helvetica,arial,freesans,cle=
an,sans-serif;font-size:13px;line-height:20px">Regex support (turned on by =
the \r escape sequence)</span></li>
<li><span style=3D"color:rgb(51,51,51);font-family:helvetica,arial,freesans=
,clean,sans-serif;font-size:13px;line-height:20px">Smartcase searching is t=
he default now (case sensitivity is triggered by uppercase characters)</spa=
n></li>
<li><span style=3D"color:rgb(51,51,51);font-family:helvetica,arial,freesans=
,clean,sans-serif;font-size:13px;line-height:20px">Case sensitivity can be =
enforced using the \I escape sequence</span></li><li><span style=3D"color:r=
gb(51,51,51);font-family:helvetica,arial,freesans,clean,sans-serif;font-siz=
e:13px;line-height:20px">Find queries are now global across tabs and persis=
tent across browser sessions</span></li>
<li><span style=3D"color:rgb(51,51,51);font-family:helvetica,arial,freesans=
,clean,sans-serif;font-size:13px;line-height:20px">Page shortcuts e.g. on G=
itHub should no longer be triggered during find mode</span></li><li><span s=
tyle=3D"color:rgb(51,51,51);font-family:helvetica,arial,freesans,clean,sans=
-serif;font-size:13px;line-height:20px">Find mode no longer triggers insert=
mode if there are input elements that match the query, but we can enter in=
sert mode on one of the results by hitting</span><span style=3D"color:rgb(5=
1,51,51);font-family:helvetica,arial,freesans,clean,sans-serif;font-size:13=
px;line-height:20px">=A0</span><code style=3D"color:rgb(51,51,51);backgroun=
d-color:rgb(248,248,248);margin-top:0px;margin-right:2px;margin-bottom:0px;=
margin-left:2px;padding-top:0px;padding-right:5px;padding-bottom:0px;paddin=
g-left:5px;font:normal normal normal 12px/normal 'Bitstream Vera Sans M=
ono',Courier,monospace;white-space:nowrap;border-top-width:1px;border-r=
ight-width:1px;border-bottom-width:1px;border-left-width:1px;border-top-sty=
le:solid;border-right-style:solid;border-bottom-style:solid;border-left-sty=
le:solid;border-top-color:rgb(234,234,234);border-right-color:rgb(234,234,2=
34);border-bottom-color:rgb(234,234,234);border-left-color:rgb(234,234,234)=
;border-top-left-radius:3px;border-top-right-radius:3px;border-bottom-right=
-radius:3px;border-bottom-left-radius:3px"><esc></code><span style=3D=
"color:rgb(51,51,51);font-family:helvetica,arial,freesans,clean,sans-serif;=
font-size:13px;line-height:20px">.</span></li>
</ul><li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean,=
sans-serif"><span style=3D"line-height:20px">Insert mode is automatically =
exited when changing tabs</span></font></li><li><font color=3D"#333333" fac=
e=3D"helvetica, arial, freesans, clean, sans-serif"><span style=3D"line-hei=
ght:20px">Many many many bugfixes and minor improvements.</span></font></li=
>
</ul><div><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean=
, sans-serif"><span style=3D"line-height:20px">Things I'd like to see s=
ettled before the push:</span></font></div><div><ul><li><font color=3D"#333=
333" face=3D"helvetica, arial, freesans, clean, sans-serif"><span style=3D"=
line-height:20px">Escape in insert mode should not be passed to page -- <a =
href=3D"https://github.com/philc/vimium/issues/468" target=3D"_blank">https=
://github.com/philc/vimium/issues/468</a> (Waiting for Ilya's input on =
this -- If you don't remember the reason for the original commit, could=
we just back it out?)</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Have escape focus content -- <a hr=
ef=3D"https://github.com/philc/vimium/issues/138" target=3D"_blank">https:/=
/github.com/philc/vimium/issues/138</a> -- current implementation / interfa=
ce is awkward, I plan to back it out before the release</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">Find mode behavior needs to be doc=
umented -- I'm thinking of linking to a separate HTML page from our hel=
p dialog</span></font></li>
<li><font color=3D"#333333" face=3D"helvetica, arial, freesans, clean, sans=
-serif"><span style=3D"line-height:20px">I don't like `g0` and `g$`. g$=
in particular seems *really* awkward to press. Could we do simple tab numb=
ering instead? I.e. g1 for the first tab, g2 for the second,, and g0 for th=
e last tab. We can also add in negative numbers -- `g-1` for the second tab=
from the end, etc. It's pretty similar to Chrome / FF's default Ct=
rl/Cmd-[number] behavior for switching tabs, but a) we'll have negative=
numbers and b) `g` is considerably less awkward than `Ctrl`, which would b=
enefit our non-Mac users. Or we could back this out as well and do some exp=
erimentation before our next release.</span></font></li>
<span><font color=3D"#888888">
</font></span></ul></div><span><font color=3D"#888888">Jez<br>
</font></span></blockquote></div><br></div></div></div>
</blockquote></div><br></div></div></div></div></div>
</blockquote></div><br></div></div>
</div></div></blockquote></div><br></div></div></div>
</blockquote></div><br></div></div>
--bcaec52be479ab051404b908c3c6--