October 2016 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Again about scrolling with a trackpad

by Yuriy Tymchuk

I have also problem in Finder when I scroll results for exampleâ¦ > On 21 Oct 2016, at 16:08, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > > Hi Yuriy. > > I just want to say that I feel completely same. > And I am sure that problem with Nautilus. I am working on remote browser where I just use FastTable and never saw any problem with scrolling. > > 2016-10-21 15:56 GMT+02:00 Yuriy Tymchuk <yuriy.tymchuk(a)me.com <mailto:yuriy.tymchuk@me.com>>: > Ok, > > there is a problem that when you use trackpad with pharo and you scroll something strange happens. And it was documented that when you scroll sideways itâs registered as ctrl + arrow click and it switches focus. > > Now it seems that on Siera it became worse because now I cannot normally scroll at all. > > Additionally this causes a lot of problems: i.e. when someone wanted to remove history plugin from nautilus many of people protested because when your navigation is fucked up during scrolling you can get back. > > Now I started to think: why wonât we simply delete the code that registers these events. So I have a question, does anybody find this functionality useful? I find it super frustrating. And if everyone else does, maybe we can simply remove that. > > Also while the time passes, does anybody have any idea on how to hack Pharo to ignore these events? Because I need to work and I can notâ¦ > > Cheers. > Uko >

Oct. 21, 2016

Re: [Pharo-dev] [ANN] Sparta v1.1

by Tudor Girba

Hi, > On Oct 21, 2016, at 3:55 PM, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > > > 2016-10-21 15:21 GMT+02:00 Tudor Girba <tudor(a)tudorgirba.com>: > > but at the rate we > > reinvent the wheel, I often fear that I will be retired from programming > > before we have a sane text model :/ > > I would kindly ask for patience. What is perhaps less clear is that this editor is in the critical path of the GT project and we are committed to deliver an editor that actually works. We are still investigating different paths, both on the low level (like with Rope) and at the higher level (like how to organize the layout). > > I am not against to any new implementation of existing stuff. It is always interesting how things could be done different and especially if it provides better solution. > But I really fear that new text experiments will dramatically delay releaze of Bloc and Sparta and following migration to new UI. And this is real importance for Pharo future, and not possible moldable editor. > Text editors are very complex domain. It takes more than year to get working TxText and Twisty. Why not finish Bloc and Sparta with minimal effort on adopting TxText or Twisty to run on them? I think there is a confusion here. Sparta was just released on all three OSs (a major effort of Alex), Bloc is working and it even has TxText working there on top of Sparta (another major effort of Glenn). As to the importance of a moldable text editor that also scales properly, I disagree with you. We need an infrastructure on top of which we can build the next 5 years of tools. Especially in the context of GT, the goal is not to reproduce the existing tools, but to create completely new ones that are not possible right now, and a scalable & moldable text editor and model is a key component for what we want to do next. That is why we will invest the effort in this part. We might fail in the process, but then again we might succeed. We can only invent a new future if we first dream of it, and then try :). Doru -- www.tudorgirba.com www.feenk.com âSoftware has no shape. Actually, it has no one shape. It has many."

Oct. 21, 2016

Re: [Pharo-dev] [ANN] Sparta v1.1

by Dale Henrichs

Norbert, I didn't realize that you were claiming that the new text model for Sparta was (potentially) inferior. The other day you were expressing sadness about having to use the newer version of Metacello (which is *only* 3 years old), so I assumed that you were just being generally cranky about change:). Dale On 10/21/16 6:50 AM, Norbert Hartl wrote: > Dale, > > I was not arguing against evolution. I was refering to the fact that > work of others is mostly ignored just to come up with a potential > weaker solution. > > Norbert > >> Am 21.10.2016 um 15:34 schrieb Dale Henrichs >> <dale.henrichs(a)gemtalksystems.com >> <mailto:dale.henrichs@gemtalksystems.com>>: >> >> Norbert, >> >> It is also called evolution ... with each step forward new >> possibilities are revealed and often the old (current) way of doing >> things needs to change to better leverage these new possibilities ... >> and evolution is required[4]. >> >> Change is also painful and keeping up with an ever-changing system >> takes a lot of individual effort. >> >> If you don't want to deal with change, then pick a Pharo version and >> stick with it ... Pharo3.0 is still functional and I for one make >> sure that all new versions of Metacello work on older versions of >> Squeak, Pharo and GemStone[1] --- but you don't have to use the newer >> version of Metacello if you don't want to:). >> >> I use Pharo3.0 on a daily basis for tODE -- I decided that I wanted >> to spend my time evolving and changing tODE itself rather than spend >> a portion of every year porting to a newer version of Pharo. >> Fortunately I don't NEED the fancy new widgets in Pharo3.0 to make >> progress with tODE. >> >> Metacello is not the only project to ensure that it continues to >> function on older versions of Pharo. Seaside[2], Voyage[3] and I'm >> sure others make an effort to continue to function on older versions >> of Pharo --- the technology exists for maintaining compatibility with >> older versions of Pharo for those projects that aren't making the >> effort now. >> >> If you choose to leverage the benefits of the newer versions of >> Pharo, then you must accept the cost, but you can pace yourself if >> the rate of change becomes too much. >> >> Dale >> >> [1] https://travis-ci.org/dalehenrich/metacello-work/builds/168940183 >> >> [2] https://travis-ci.org/SeasideSt/Seaside/builds/160382244 >> >> [3] https://travis-ci.org/pharo-nosql/voyage/builds/167012791 >> >> [4] https://en.wikipedia.org/wiki/Software_evolution >> >> >> On 10/21/16 12:10 AM, Norbert Hartl wrote: >>> >>> >>> Am 21.10.2016 um 04:00 schrieb Sean P. DeNigris >>> <sean(a)clipperadams.com <mailto:sean@clipperadams.com>>: >>> >>>> Denis Kudriashov wrote >>>>> I look at code and it seems you implemented another one new text >>>>> model? >>>>> Why >>>>> you not use TxText? >>>> >>>> Argh. I know it's bad form to complain about gifts, but at the rate we >>>> reinvent the wheel, I often fear that I will be retired from >>>> programming >>>> before we have a sane text model :/ >>>> >>> It has a name and we should fight it: >>> >>> https://en.m.wikipedia.org/wiki/Not_invented_here >>> >>> Norbert >>>> >>>> ----- >>>> Cheers, >>>> Sean >>>> -- >>>> View this message in context: >>>> http://forum.world.st/ANN-Sparta-v1-1-tp4919394p4919570.html >>>> Sent from the Pharo Smalltalk Developers mailing list archive at >>>> Nabble.com <http://nabble.com/>. >>>> >> >

Oct. 21, 2016

Re: [Pharo-dev] Again about scrolling with a trackpad

by Denis Kudriashov

Hi Yuriy. I just want to say that I feel completely same. And I am sure that problem with Nautilus. I am working on remote browser where I just use FastTable and never saw any problem with scrolling. 2016-10-21 15:56 GMT+02:00 Yuriy Tymchuk <yuriy.tymchuk(a)me.com>: > Ok, > > there is a problem that when you use trackpad with pharo and you scroll > something strange happens. And it was documented that when you scroll > sideways itâs registered as ctrl + arrow click and it switches focus. > > Now it seems that on Siera it became worse because now I cannot normally > scroll at all. > > Additionally this causes a lot of problems: i.e. when someone wanted to > remove history plugin from nautilus many of people protested because when > your navigation is fucked up during scrolling you can get back. > > Now I started to think: why wonât we simply delete the code that registers > these events. So I have a question, does anybody find this functionality > useful? I find it super frustrating. And if everyone else does, maybe we > can simply remove that. > > Also while the time passes, does anybody have any idea on how to hack > Pharo to ignore these events? Because I need to work and I can notâ¦ > > Cheers. > Uko >

Oct. 21, 2016

Re: [Pharo-dev] Git versions on CI slaves

by Damien Pollet

On 21 October 2016 at 15:47, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > The main problem are the long-term supported Linux distributions like > Debian or CentOS. Even Debian stable has git 2.1 so I'd say it's a non-problem. -- Damien Pollet type less, do more [ | ] http://people.untyped.org/damien.pollet

Oct. 21, 2016

Re: [Pharo-dev] [ANN] Sparta v1.1

by Tudor Girba

Hi Norbert, It is not ignored at all. Cheers, Doru > On Oct 21, 2016, at 3:50 PM, Norbert Hartl <norbert(a)hartl.name> wrote: > > Dale, > > I was not arguing against evolution. I was refering to the fact that work of others is mostly ignored just to come up with a potential weaker solution. > > Norbert > >> Am 21.10.2016 um 15:34 schrieb Dale Henrichs <dale.henrichs(a)gemtalksystems.com>: >> >> Norbert, >> >> It is also called evolution ... with each step forward new possibilities are revealed and often the old (current) way of doing things needs to change to better leverage these new possibilities ... and evolution is required[4]. >> Change is also painful and keeping up with an ever-changing system takes a lot of individual effort. >> >> If you don't want to deal with change, then pick a Pharo version and stick with it ... Pharo3.0 is still functional and I for one make sure that all new versions of Metacello work on older versions of Squeak, Pharo and GemStone[1] --- but you don't have to use the newer version of Metacello if you don't want to:). >> >> I use Pharo3.0 on a daily basis for tODE -- I decided that I wanted to spend my time evolving and changing tODE itself rather than spend a portion of every year porting to a newer version of Pharo. Fortunately I don't NEED the fancy new widgets in Pharo3.0 to make progress with tODE. >> Metacello is not the only project to ensure that it continues to function on older versions of Pharo. Seaside[2], Voyage[3] and I'm sure others make an effort to continue to function on older versions of Pharo --- the technology exists for maintaining compatibility with older versions of Pharo for those projects that aren't making the effort now. >> If you choose to leverage the benefits of the newer versions of Pharo, then you must accept the cost, but you can pace yourself if the rate of change becomes too much. >> >> Dale >> [1] https://travis-ci.org/dalehenrich/metacello-work/builds/168940183 >> >> [2] https://travis-ci.org/SeasideSt/Seaside/builds/160382244 >> >> [3] https://travis-ci.org/pharo-nosql/voyage/builds/167012791 >> [4] https://en.wikipedia.org/wiki/Software_evolution >> >> On 10/21/16 12:10 AM, Norbert Hartl wrote: >>> >>> >>> Am 21.10.2016 um 04:00 schrieb Sean P. DeNigris <sean(a)clipperadams.com>: >>> >>>> Denis Kudriashov wrote >>>>> I look at code and it seems you implemented another one new text model? >>>>> Why >>>>> you not use TxText? >>>> >>>> Argh. I know it's bad form to complain about gifts, but at the rate we >>>> reinvent the wheel, I often fear that I will be retired from programming >>>> before we have a sane text model :/ >>>> >>> It has a name and we should fight it: >>> >>> https://en.m.wikipedia.org/wiki/Not_invented_here >>> >>> Norbert >>>> >>>> ----- >>>> Cheers, >>>> Sean >>>> -- >>>> View this message in context: http://forum.world.st/ANN-Sparta-v1-1-tp4919394p4919570.html >>>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>>> >> > -- www.tudorgirba.com www.feenk.com "Some battles are better lost than fought."

Oct. 21, 2016

Again about scrolling with a trackpad

by Yuriy Tymchuk

Ok, there is a problem that when you use trackpad with pharo and you scroll something strange happens. And it was documented that when you scroll sideways itâs registered as ctrl + arrow click and it switches focus. Now it seems that on Siera it became worse because now I cannot normally scroll at all. Additionally this causes a lot of problems: i.e. when someone wanted to remove history plugin from nautilus many of people protested because when your navigation is fucked up during scrolling you can get back. Now I started to think: why wonât we simply delete the code that registers these events. So I have a question, does anybody find this functionality useful? I find it super frustrating. And if everyone else does, maybe we can simply remove that. Also while the time passes, does anybody have any idea on how to hack Pharo to ignore these events? Because I need to work and I can notâ¦ Cheers. Uko

Oct. 21, 2016

Re: [Pharo-dev] [ANN] Sparta v1.1

by Denis Kudriashov

2016-10-21 15:21 GMT+02:00 Tudor Girba <tudor(a)tudorgirba.com>: > > but at the rate we > > reinvent the wheel, I often fear that I will be retired from programming > > before we have a sane text model :/ > > I would kindly ask for patience. What is perhaps less clear is that this > editor is in the critical path of the GT project and we are committed to > deliver an editor that actually works. We are still investigating different > paths, both on the low level (like with Rope) and at the higher level (like > how to organize the layout). > I am not against to any new implementation of existing stuff. It is always interesting how things could be done different and especially if it provides better solution. But I really fear that new text experiments will dramatically delay releaze of Bloc and Sparta and following migration to new UI. And this is real importance for Pharo future, and not possible moldable editor. Text editors are very complex domain. It takes more than year to get working TxText and Twisty. Why not finish Bloc and Sparta with minimal effort on adopting TxText or Twisty to run on them?

Oct. 21, 2016

Re: [Pharo-dev] [ANN] Sparta v1.1

by Norbert Hartl

Dale, I was not arguing against evolution. I was refering to the fact that work of others is mostly ignored just to come up with a potential weaker solution. Norbert > Am 21.10.2016 um 15:34 schrieb Dale Henrichs <dale.henrichs(a)gemtalksystems.com>: > > Norbert, > > It is also called evolution ... with each step forward new possibilities are revealed and often the old (current) way of doing things needs to change to better leverage these new possibilities ... and evolution is required[4]. > Change is also painful and keeping up with an ever-changing system takes a lot of individual effort. > > If you don't want to deal with change, then pick a Pharo version and stick with it ... Pharo3.0 is still functional and I for one make sure that all new versions of Metacello work on older versions of Squeak, Pharo and GemStone[1] --- but you don't have to use the newer version of Metacello if you don't want to:). > > I use Pharo3.0 on a daily basis for tODE -- I decided that I wanted to spend my time evolving and changing tODE itself rather than spend a portion of every year porting to a newer version of Pharo. Fortunately I don't NEED the fancy new widgets in Pharo3.0 to make progress with tODE. > Metacello is not the only project to ensure that it continues to function on older versions of Pharo. Seaside[2], Voyage[3] and I'm sure others make an effort to continue to function on older versions of Pharo --- the technology exists for maintaining compatibility with older versions of Pharo for those projects that aren't making the effort now. > If you choose to leverage the benefits of the newer versions of Pharo, then you must accept the cost, but you can pace yourself if the rate of change becomes too much. > > Dale > [1] https://travis-ci.org/dalehenrich/metacello-work/builds/168940183 <https://travis-ci.org/dalehenrich/metacello-work/builds/168940183> > [2] https://travis-ci.org/SeasideSt/Seaside/builds/160382244 <https://travis-ci.org/SeasideSt/Seaside/builds/160382244> > [3] https://travis-ci.org/pharo-nosql/voyage/builds/167012791 <https://travis-ci.org/pharo-nosql/voyage/builds/167012791> > [4] https://en.wikipedia.org/wiki/Software_evolution <https://en.wikipedia.org/wiki/Software_evolution> > > On 10/21/16 12:10 AM, Norbert Hartl wrote: >> >> >> Am 21.10.2016 um 04:00 schrieb Sean P. DeNigris <sean(a)clipperadams.com <mailto:sean@clipperadams.com>>: >> >>> Denis Kudriashov wrote >>>> I look at code and it seems you implemented another one new text model? >>>> Why >>>> you not use TxText? >>> >>> Argh. I know it's bad form to complain about gifts, but at the rate we >>> reinvent the wheel, I often fear that I will be retired from programming >>> before we have a sane text model :/ >>> >> It has a name and we should fight it: >> >> https://en.m.wikipedia.org/wiki/Not_invented_here <https://en.m.wikipedia.org/wiki/Not_invented_here> >> >> Norbert >>> >>> ----- >>> Cheers, >>> Sean >>> -- >>> View this message in context: http://forum.world.st/ANN-Sparta-v1-1-tp4919394p4919570.html <http://forum.world.st/ANN-Sparta-v1-1-tp4919394p4919570.html> >>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com <http://nabble.com/>. >>> >

Oct. 21, 2016

Re: [Pharo-dev] Git versions on CI slaves

by Pavel Krivanek

The -C option changes the Git working directory and it is used in GitFileTree on general level. The main question that we need to answer is what version of Git (or libgit) will be supported and then take decision if we should update slaves or modify our code. The main problem are the long-term supported Linux distributions like Debian or CentOS. -- Pavel 2016-10-21 15:39 GMT+02:00 Nicolas Passerini <npasserini(a)gmail.com>: > So what is the path to go? > Shall we update the slaves to use the newer version or should we see how > to remove that option? > > BTW, the -C option is in which git command? > > 2016-10-21 11:20 GMT+02:00 Pavel Krivanek <pavel.krivanek(a)gmail.com>: > >> Hi, >> >> we have a problem with current MonticelloFileTree (and Iceberg) on our CI >> slaves because it requires option -C that was introduced in version 1.8.5 >> or something like so but our slaves have older Git versions (1.7). >> >> -- Pavel >> > >

Oct. 21, 2016