We all know (and lament) that QS is no longer being developed. In an
interview some time ago, Alcor was even recommending that folks
switch. Inevitably, the lack of further development (or maintenance)
means that sometime in the future a software update will break it.
I was an early user of Launchbar way back when, but when QS came out
of Blacktree, I was hooked -- binned LB and used QS ever since.
Searching here, I see nothing but negatives about the current LB. At
the same time, QS is not sufficiently "Apple-like" to appeal to them;
it has a learning curve and Apple doesn't like learning curves. Is
there no hope that some member of the consortium of developers that
worked with Alcor has any ambition to continue? Is QS really going to
remain dead in the water?
On May 31, 8:13 pm, NovaScotian <ACB...@gmail.com> wrote:
> it has a learning curve and Apple doesn't like learning curves. Is
> there no hope that some member of the consortium of developers that
> worked with Alcor has any ambition to continue? Is QS really going to
> remain dead in the water?
It really surprises me that no one has stepped up. QS has one of the
most dedicated and fanatical userbases (steadily eroding) of any Mac
software, and is one of the most brilliant thought out interfaces ever
on any platform (talked about enviously by Windows and *nix users
alike). It has been opensourced and sits in a repository for someone
to hack on. Many of us are willing to give *very* generously to a tip
jar, and thus there is a financial incentive that is probably greater
than many other opensource projects combined (though there are some
ethical issues involved there).
I think devs don't touch it because it is considered a dead branch
with Alcor working on a new core (i.e. they don't want to step on his
shoes). However that work is stalled (no activity in Google code for
months), and thus the whole endeavour is frozen in stasis. No one
wants to waste time on a dead branch, and yet the new tree is frozen
in its growth.
We have to be grateful to alcor (and the legion of QS users who
contributed to its development, and people like Howard for his amazing
documentation). It is really sad to see such energy slowly dissipate
away, leaving us with broken brilliance.
My only hope is that another young gun takes the ideas contained
within quicksilver, along with new ideas and amkes something even
better. Alternatively Google gives Alcor a 6 month sabbatical for the
good of OS X and makes QS II part of its stable of tools.
Does anyone have any information about the current state of both branches? Both source locations as well as what's broken ?
I wouldn't mind taking a look to see the state of things and see just how bad everything is.
It seems as though "Stepping on Alcors shoes" is almost a non- possibility at this moment because he doesn't have the interest / time to take QS further.
- Griffin On Jun 1, 2008, at 3:01 AM, nontoppo wrote:
> On May 31, 8:13 pm, NovaScotian <ACB...@gmail.com> wrote:
>> it has a learning curve and Apple doesn't like learning curves. Is >> there no hope that some member of the consortium of developers that >> worked with Alcor has any ambition to continue? Is QS really going to >> remain dead in the water?
> It really surprises me that no one has stepped up. QS has one of the > most dedicated and fanatical userbases (steadily eroding) of any Mac > software, and is one of the most brilliant thought out interfaces ever > on any platform (talked about enviously by Windows and *nix users > alike). It has been opensourced and sits in a repository for someone > to hack on. Many of us are willing to give *very* generously to a tip > jar, and thus there is a financial incentive that is probably greater > than many other opensource projects combined (though there are some > ethical issues involved there).
> I think devs don't touch it because it is considered a dead branch > with Alcor working on a new core (i.e. they don't want to step on his > shoes). However that work is stalled (no activity in Google code for > months), and thus the whole endeavour is frozen in stasis. No one > wants to waste time on a dead branch, and yet the new tree is frozen > in its growth.
> We have to be grateful to alcor (and the legion of QS users who > contributed to its development, and people like Howard for his amazing > documentation). It is really sad to see such energy slowly dissipate > away, leaving us with broken brilliance.
> My only hope is that another young gun takes the ideas contained > within quicksilver, along with new ideas and amkes something even > better. Alternatively Google gives Alcor a 6 month sabbatical for the > good of OS X and makes QS II part of its stable of tools.
QS II on trunk (for want of a title) had a heap of activity over
christmas. It is a restructuring of QS, from the readme:
Trunk contains the source for the experimental version of Quicksilver.
It differs from the released version in the following ways:
* Triggers are moving to a separate product, called Catalyst
* All the little frameworks are being joined into one big one called
Crucible. This includes extensions and core functionality that most
apps and plugins will use. This is currently called QSBase.framework
* The preferences are going to get MUCH simpler. There will be
Extras-style advanced prefs for the fiddly options.
* Plugins are going to be hidden from most users, they'll activate
themselves automatically or be installable from the web
* B5X Plugins are incompatible, mostly because that architecture was
a mess.
--------------------------------
Crucible
A framework with extension to AppKit and tools common to all Alchemy
apps
Elements
A framework supporting the plugin architecture
Quicksilver
Command window driven launcher
Catalyst
Triggers preference pane
B5X is the branch which the curent QS is built from. Ankur did some
work on it (a build off rev76) and has claimed he has done a bunch
more (reducing QS's memory consumption, though hasn't merged it yet),
though has no time until the end of this year to take things further:
That branch is working though with the leopard bugs and crashes we've
all noted. Fixing up the leopard bugs seems entirely reasonable with
B5X and would be an enormous boost for all QS users.
> It seems as though "Stepping on Alcors shoes" is almost a non-
> possibility at this moment because he doesn't have the interest / time
> to take QS further.
No idea, there was lots of activity on QS II and we don't know if it
will pick up again as he hasn't said one way or the other.
Rather than working on the B5X branch (depends on Alcor to accept your
work?), I think people should just branch off and do what they can
(ideally several devs working together). A legion of QS users will be
in your debt (and willing to beta test extensively) :-)
Thanks for the points. I'm checking out the source code now and I'll take a look. My biggest reservation is the fact that I don't know much about the dev environment QS was written in. I'll poke around and see if I can figure anything out.
It would be nice if I could at least get a local build running on my box(es).
- Griffin On Jun 2, 2008, at 2:03 AM, nontoppo wrote:
> On Jun 2, 1:51 am, Griffin Caprio <griffin.cap...@gmail.com> wrote: >> Does anyone have any information about the current state of both >> branches? Both source locations as well as what's broken ?
> QS II on trunk (for want of a title) had a heap of activity over > christmas. It is a restructuring of QS, from the readme:
> Trunk contains the source for the experimental version of Quicksilver. > It differs from the released version in the following ways: > * Triggers are moving to a separate product, called Catalyst > * All the little frameworks are being joined into one big one called > Crucible. This includes extensions and core functionality that most > apps and plugins will use. This is currently called QSBase.framework > * The preferences are going to get MUCH simpler. There will be > Extras-style advanced prefs for the fiddly options. > * Plugins are going to be hidden from most users, they'll activate > themselves automatically or be installable from the web > * B5X Plugins are incompatible, mostly because that architecture was > a mess. > -------------------------------- > Crucible > A framework with extension to AppKit and tools common to all Alchemy > apps > Elements > A framework supporting the plugin architecture > Quicksilver > Command window driven launcher > Catalyst > Triggers preference pane
> B5X is the branch which the curent QS is built from. Ankur did some > work on it (a build off rev76) and has claimed he has done a bunch > more (reducing QS's memory consumption, though hasn't merged it yet), > though has no time until the end of this year to take things further:
> That branch is working though with the leopard bugs and crashes we've > all noted. Fixing up the leopard bugs seems entirely reasonable with > B5X and would be an enormous boost for all QS users.
>> It seems as though "Stepping on Alcors shoes" is almost a non- >> possibility at this moment because he doesn't have the interest / >> time >> to take QS further.
> No idea, there was lots of activity on QS II and we don't know if it > will pick up again as he hasn't said one way or the other.
> Rather than working on the B5X branch (depends on Alcor to accept your > work?), I think people should just branch off and do what they can > (ideally several devs working together). A legion of QS users will be > in your debt (and willing to beta test extensively) :-)
On Mon, Jun 2, 2008 at 12:03 AM, nontoppo <ian...@gmail.com> wrote:
> Trunk contains the source for the experimental version of Quicksilver. > It differs from the released version in the following ways: > * Triggers are moving to a separate product, called Catalyst > * All the little frameworks are being joined into one big one called > Crucible. This includes extensions and core functionality that most > apps and plugins will use. This is currently called QSBase.framework > * The preferences are going to get MUCH simpler. There will be > Extras-style advanced prefs for the fiddly options. > * Plugins are going to be hidden from most users, they'll activate > themselves automatically or be installable from the web > * B5X Plugins are incompatible, mostly because that architecture was > a mess. [...]
This sounds like a Big Bang rearchitecture that is so rarely successful. What tends to be successful (and more useful to us users) is to figure out what the users need (as opposed to what the developer likes to work on), release that, and repeat. If this rearchitecture ever is released, it's going to be so buggy (because no real people have ever used so much of its code) that people will stay away in droves.
Just a prediction based on experience. I hope it doesn't come true.
On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote:
> This sounds like a Big Bang rearchitecture that is so rarely successful.
Quite likely to come true.
At this point I'd like to make a recommendation:
It is probably better to just release the branch. It is actually quite
close to being ready. With QuickLook and some of the bug fixes, would
be substantially better than the current release.
I don't want to encourage too much work on QS because (shortly) I
intend to make it obsolete, but giving users something that will work
for the next year is certainly worthwhile.
> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote:
> > This sounds like a Big Bang rearchitecture that is so rarely successful.
> Quite likely to come true.
> At this point I'd like to make a recommendation:
> It is probably better to just release the branch. It is actually quite
> close to being ready. With QuickLook and some of the bug fixes, would
> be substantially better than the current release.
> I don't want to encourage too much work on QS because (shortly) I
> intend to make it obsolete, but giving users something that will work
> for the next year is certainly worthwhile.
That's great news to hear. Can I ask a couple of questions:
1 ) Who would be in charge of "releasing the branch" as you say? 2 ) What needs to be done in order for the branch to be released? Documentation? Beta Testing ?
> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote: >> This sounds like a Big Bang rearchitecture that is so rarely >> successful.
> Quite likely to come true.
> At this point I'd like to make a recommendation:
> It is probably better to just release the branch. It is actually quite > close to being ready. With QuickLook and some of the bug fixes, would > be substantially better than the current release.
> I don't want to encourage too much work on QS because (shortly) I > intend to make it obsolete, but giving users something that will work > for the next year is certainly worthwhile.
Great. I guess the only question for people now is do they want to continue using a product knowing it will be made obsolete. I'm fine using a new, leopard compatible build for at least the next year. So for me, that's a yes.
- Griffin On Jun 3, 2008, at 9:06 AM, J. Nicholas Jitkoff wrote:
> I can release it, mostly we just need people trying it out to see if > it feels ok. I have a few outstanding changes to add quick look, and > I'll need to put those in.
> On Jun 3, 2008, at 6:58 AM, Griffin Caprio wrote:
>> Hi Alcor,
>> That's great news to hear. Can I ask a couple of questions:
>> 1 ) Who would be in charge of "releasing the branch" as you say? >> 2 ) What needs to be done in order for the branch to be released? >> Documentation? Beta Testing ?
>> Thanks, >> Griffin Caprio
>> On Jun 3, 2008, at 8:03 AM, AIcor wrote:
>>> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote: >>>> This sounds like a Big Bang rearchitecture that is so rarely >>>> successful.
>>> Quite likely to come true.
>>> At this point I'd like to make a recommendation:
>>> It is probably better to just release the branch. It is actually >>> quite >>> close to being ready. With QuickLook and some of the bug fixes, >>> would >>> be substantially better than the current release.
>>> I don't want to encourage too much work on QS because (shortly) I >>> intend to make it obsolete, but giving users something that will >>> work >>> for the next year is certainly worthwhile.
> I'll have something better for you by the time Leopard is done for.
> On Jun 3, 2008, at 7:10 AM, Griffin Caprio wrote:
>> Great. I guess the only question for people now is do they want to >> continue using a product knowing it will be made obsolete. I'm >> fine using a new, leopard compatible build for at least the next >> year. So for me, that's a yes.
>> - Griffin >> On Jun 3, 2008, at 9:06 AM, J. Nicholas Jitkoff wrote:
>>> I can release it, mostly we just need people trying it out to see >>> if it feels ok. I have a few outstanding changes to add quick >>> look, and I'll need to put those in.
>>> On Jun 3, 2008, at 6:58 AM, Griffin Caprio wrote:
>>>> Hi Alcor,
>>>> That's great news to hear. Can I ask a couple of questions:
>>>> 1 ) Who would be in charge of "releasing the branch" as you say? >>>> 2 ) What needs to be done in order for the branch to be >>>> released? Documentation? Beta Testing ?
>>>> Thanks, >>>> Griffin Caprio
>>>> On Jun 3, 2008, at 8:03 AM, AIcor wrote:
>>>>> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote: >>>>>> This sounds like a Big Bang rearchitecture that is so rarely >>>>>> successful.
>>>>> Quite likely to come true.
>>>>> At this point I'd like to make a recommendation:
>>>>> It is probably better to just release the branch. It is actually >>>>> quite >>>>> close to being ready. With QuickLook and some of the bug fixes, >>>>> would >>>>> be substantially better than the current release.
>>>>> I don't want to encourage too much work on QS because (shortly) I >>>>> intend to make it obsolete, but giving users something that will >>>>> work >>>>> for the next year is certainly worthwhile.
> In the mean time, i'd be all for testing a new build of the branch.
> Any idea on if Ankur's changed would be merged in?
> - Griffin
> On Jun 3, 2008, at 9:17 AM, J. Nicholas Jitkoff wrote:
> > I'll have something better for you by the time Leopard is done for.
> > On Jun 3, 2008, at 7:10 AM, Griffin Caprio wrote:
> >> Great. I guess the only question for people now is do they want to
> >> continue using a product knowing it will be made obsolete. I'm
> >> fine using a new, leopard compatible build for at least the next
> >> year. So for me, that's a yes.
> >> - Griffin
> >> On Jun 3, 2008, at 9:06 AM, J. Nicholas Jitkoff wrote:
> >>> I can release it, mostly we just need people trying it out to see
> >>> if it feels ok. I have a few outstanding changes to add quick
> >>> look, and I'll need to put those in.
> >>> On Jun 3, 2008, at 6:58 AM, Griffin Caprio wrote:
> >>>> Hi Alcor,
> >>>> That's great news to hear. Can I ask a couple of questions:
> >>>> 1 ) Who would be in charge of "releasing the branch" as you say?
> >>>> 2 ) What needs to be done in order for the branch to be
> >>>> released? Documentation? Beta Testing ?
> >>>> Thanks,
> >>>> Griffin Caprio
> >>>> On Jun 3, 2008, at 8:03 AM, AIcor wrote:
> >>>>> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote:
> >>>>>> This sounds like a Big Bang rearchitecture that is so rarely
> >>>>>> successful.
> >>>>> Quite likely to come true.
> >>>>> At this point I'd like to make a recommendation:
> >>>>> It is probably better to just release the branch. It is actually
> >>>>> quite
> >>>>> close to being ready. With QuickLook and some of the bug fixes,
> >>>>> would
> >>>>> be substantially better than the current release.
> >>>>> I don't want to encourage too much work on QS because (shortly) I
> >>>>> intend to make it obsolete, but giving users something that will
> >>>>> work
> >>>>> for the next year is certainly worthwhile.
> this is great news to hear. I would also like to know if ankur's
> updates will be merged.
> Alcor, care to elaborate a tad on the new project?
> On Jun 3, 7:30 am, Griffin Caprio <griffin.cap...@gmail.com> wrote:
> > Wow, that's great to hear.
> > In the mean time, i'd be all for testing a new build of the branch.
> > Any idea on if Ankur's changed would be merged in?
> > - Griffin
> > On Jun 3, 2008, at 9:17 AM, J. Nicholas Jitkoff wrote:
> > > I'll have something better for you by the time Leopard is done for.
> > > On Jun 3, 2008, at 7:10 AM, Griffin Caprio wrote:
> > >> Great. I guess the only question for people now is do they want to
> > >> continue using a product knowing it will be made obsolete. I'm
> > >> fine using a new, leopard compatible build for at least the next
> > >> year. So for me, that's a yes.
> > >> - Griffin
> > >> On Jun 3, 2008, at 9:06 AM, J. Nicholas Jitkoff wrote:
> > >>> I can release it, mostly we just need people trying it out to see
> > >>> if it feels ok. I have a few outstanding changes to add quick
> > >>> look, and I'll need to put those in.
> > >>> On Jun 3, 2008, at 6:58 AM, Griffin Caprio wrote:
> > >>>> Hi Alcor,
> > >>>> That's great news to hear. Can I ask a couple of questions:
> > >>>> 1 ) Who would be in charge of "releasing the branch" as you say?
> > >>>> 2 ) What needs to be done in order for the branch to be
> > >>>> released? Documentation? Beta Testing ?
> > >>>> Thanks,
> > >>>> Griffin Caprio
> > >>>> On Jun 3, 2008, at 8:03 AM, AIcor wrote:
> > >>>>> On Jun 2, 11:14 am, "Mark Wilden" <m...@mwilden.com> wrote:
> > >>>>>> This sounds like a Big Bang rearchitecture that is so rarely
> > >>>>>> successful.
> > >>>>> Quite likely to come true.
> > >>>>> At this point I'd like to make a recommendation:
> > >>>>> It is probably better to just release the branch. It is actually
> > >>>>> quite
> > >>>>> close to being ready. With QuickLook and some of the bug fixes,
> > >>>>> would
> > >>>>> be substantially better than the current release.
> > >>>>> I don't want to encourage too much work on QS because (shortly) I
> > >>>>> intend to make it obsolete, but giving users something that will
> > >>>>> work
> > >>>>> for the next year is certainly worthwhile.
i am willing to leave QS running as is without upgrades as it works
well for me and if there is no assurance that it will continue to do
so after tweaking. i am currently running b53 (3814) and so far so
good but i am wondering how long it would take it to get outdated. i
suspect that it should stay useful at least until leopard is current.
On Jun 2, 1:51 am, Griffin Caprio <griffin.cap...@gmail.com> wrote:
> Does anyone have any information about the current state of both
> branches? Both source locations as well as what's broken ?
Just an update: both branch and trunk are seeing quite a bit of work
on them, and there is a new wiki page on building trunk, thus I assume
it is now building properly:
Looks like the quicklook fix is going into the existing QS, most of
this is being done by tiennou7. I don't see any activity from ankur
directly, so no idea about his fixes...
> On Jun 2, 1:51 am, Griffin Caprio <griffin.cap...@gmail.com> wrote:
> > Does anyone have any information about the current state of both
> > branches? Both source locations as well as what's broken ?
> Just an update: both branch and trunk are seeing quite a bit of work
> on them, and there is a new wiki page on building trunk, thus I assume
> it is now building properly:
> Looks like the quicklook fix is going into the existing QS, most of
> this is being done by tiennou7. I don't see any activity from ankur
> directly, so no idea about his fixes...
Sorry I haven't came up earlier on the group, but I didn't even notice
the blacktree forums were down until about one week ago ;-).
I'll try smashing the bugs I can before releasing what will become QS
b55, so if you guys want to give a hand, I'll put up a QS b55a1
(scary!) on the project page, along with some instructions.
> Sorry I haven't came up earlier on the group, but I didn't even notice > the blacktree forums were down until about one week ago ;-). > I'll try smashing the bugs I can before releasing what will become QS > b55, so if you guys want to give a hand, I'll put up a QS b55a1 > (scary!) on the project page, along with some instructions.
I would be happy to do some testing and report back.
tiennou speaks :D
Count me in for testing. I'd be more than happy to help.
Talking of the Building QS wiki page, I tried to build the trunk but
it doesn't seem to want to work for me :( There are a few functions
undefined.
I'll figure it out once I've learnt Cocoa though :)
On Jun 19, 7:57 pm, "Robert Occhialini Jr." <bump...@gmail.com> wrote:
> > Sorry I haven't came up earlier on the group, but I didn't even notice
> > the blacktree forums were down until about one week ago ;-).
> > I'll try smashing the bugs I can before releasing what will become QS
> > b55, so if you guys want to give a hand, I'll put up a QS b55a1
> > (scary!) on the project page, along with some instructions.
> I would be happy to do some testing and report back.
> Sorry I haven't came up earlier on the group, but I didn't even notice > the blacktree forums were down until about one week ago ;-). > I'll try smashing the bugs I can before releasing what will become QS > b55, so if you guys want to give a hand, I'll put up a QS b55a1 > (scary!) on the project page, along with some instructions.
Right now I'm in the process of porting the build system work from the
trunk to the branch, to tackle the architecture problem once and for
all (the Notification Hub has this kind of problem).
I'll be out of touch this weekend, so I'm trying to get this alpha
build up and running so that you can extensively test it while I'm
away ;-).
Okay, I have uploaded QS at http://code.google.com/p/blacktree-alchemy/downloads/list.
I quickly tested it runs on 10.5/i386 and 10.4/ppc, and it launches on
both, but I haven't tested much more than trying to launch Console,
switching preferences tabs and enabling/disabling catalog stuff.
Time for more thorough testing guys ;-).
On 20 juin, 12:43, tiennou <tienn...@gmail.com> wrote:
> Right now I'm in the process of porting the build system work from the
> trunk to the branch, to tackle the architecture problem once and for
> all (the Notification Hub has this kind of problem).
> I'll be out of touch this weekend, so I'm trying to get this alpha
> build up and running so that you can extensively test it while I'm
> away ;-).
Can you respin the number? Someone else has released a build called
b55. There is enough confusion going around about that vs b54 from
the blacktree page. I can only imagine this will make things much
worse.
Thanks for your efforts.
On Jun 20, 7:36 am, tiennou <tienn...@gmail.com> wrote:
> Okay, I have uploaded QS athttp://code.google.com/p/blacktree-alchemy/downloads/list.
> I quickly tested it runs on 10.5/i386 and 10.4/ppc, and it launches on
> both, but I haven't tested much more than trying to launch Console,
> switching preferences tabs and enabling/disabling catalog stuff.
> Time for more thorough testing guys ;-).
> On 20 juin, 12:43, tiennou <tienn...@gmail.com> wrote:
> > Right now I'm in the process of porting the build system work from the
> > trunk to the branch, to tackle the architecture problem once and for
> > all (the Notification Hub has this kind of problem).
> > I'll be out of touch this weekend, so I'm trying to get this alpha
> > build up and running so that you can extensively test it while I'm
> > away ;-).
> Can you respin the number? Someone else has released a build called
> b55. There is enough confusion going around about that vs b54 from
> the blacktree page. I can only imagine this will make things much
> worse.
> Thanks for your efforts.
> On Jun 20, 7:36 am, tiennou <tienn...@gmail.com> wrote:
> > Okay, I have uploaded QS athttp://code.google.com/p/blacktree-alchemy/downloads/list.
> > I quickly tested it runs on 10.5/i386 and 10.4/ppc, and it launches on
> > both, but I haven't tested much more than trying to launch Console,
> > switching preferences tabs and enabling/disabling catalog stuff.
> > Time for more thorough testing guys ;-).
> > On 20 juin, 12:43, tiennou <tienn...@gmail.com> wrote:
> > > Right now I'm in the process of porting the build system work from the
> > > trunk to the branch, to tackle the architecture problem once and for
> > > all (the Notification Hub has this kind of problem).
> > > I'll be out of touch this weekend, so I'm trying to get this alpha
> > > build up and running so that you can extensively test it while I'm
> > > away ;-).
|
