June 2014 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by Tudor Girba

Hi, Thanks. Comments inline. On Mon, Jun 16, 2014 at 9:36 AM, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > > On 15 Jun 2014, at 15:04, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > > Hi, > > > > I like very much the new energy people are putting into creating the > SystemLogger engine for Pharo. I think this is a specifically important > area for which we have to have a solution out of the box. At the same time, > I also think that Pharo provides an infrastructure that makes room for > ideas that are otherwise hard to reach in other languages or environments. > > > > Stef asked for collaborations around this project, so here is my > literally small contribution: a rather different logging engine. It is > called Beacon, it is based entirely on Announcements, it has ~200 lines of > code, it has no tags or levels, and in my opinion it is fully functional. > > > > You can see a detailed description here including some informal > comparisons with SystemLogger: > > http://www.humane-assessment.com/blog/beacon > > > > Please let me know what you think. I would be happy to join forces to > reach a mature solution that is both versatile and that can show how Pharo > is different. > > Here is some of my feedback: > > - Like I said in my document and like I tried to show in my implementation > for Zinc, I want framework agnostic pure object logging, so I don't want to > subclass from anything (I do from Announcement, but that is an > implementation detail, the technique how #emit is done), can you deal with > the ZnNewLogEvents as the are ? > As you can see at the end of the post, you can just use your own Announcements. There is no real magic there. So, it should be no technical problem to accommodate the logging of ZincEvents as they are. However, one thing that we should also think about is the actual analyses that we might want to do. For example, the inspector extension for RecordingBeacon uses: - timestamp, and - printOneLineContentsOn: And these requirements will grow even more as we are building more specialized tools. So, to this end, it is useful to have some common ground. Now, as far as I know, the only reason why you do not want to subclass BeaconSignal is because you want to use ZnTimestamp. I see no reason to have two Timestamps in the image anyway, so I have no problem pushing just one. Or do you have another reason? > - I find what you did in StackSignal very interesting, when putting > Exceptions in log objects, a natural thing to do, the same problems occur. > I think we are just exploring this area and have lots to learn (debugging > in the past). > Exactly. Andrei and I already used the StackSignal for a real debugging session and I will blog about it shortly. > - The fact that FuelBeacon creates a new file for every signal does not > strike me as efficient enough for real world use. Also, using the > #printString as filename is weird for me. > As mentioned in the post, this is more of an example than something to be used for real. We definitely need real solutions in this space. > - I am also wondering whether the fact that Announcements are synchronous > (I mean distributed to consumers on the same thread as the one that > produced them) is good or not because like this, slow backend processing > for whatever reason will kill performance of the producing code. > Good point. I thought about it as well, but I do not have a good solution. It would definitely be interesting to have solutions both for sync and async. > - In your blog post (very good as usual ;-) I am still bothered by the > usage of strings in the examples (I know you also use the stack signal, but > that is more advanced). That way, reader will still think about text. We > need better examples that show the power of object logging. I'll see if I > can come up with something. > I know! It pained me as well, but I did not have good examples at hand (except for the stack one that I needed first hand) and I thought that it is better to bring it up for discussion than wait until it is too late. > - Small code change > > StringStreamBeacon>>#handle: anAnnoucement > self stream > nextPutAll: anAnnoucement printString; > cr > > would be more efficient (no intermediate string) as > > StringStreamBeacon>>#handle: anAnnoucement > self stream > print: anAnnoucement; > cr > > or even > > StringStreamBeacon>>#handle: anAnnoucement > anAnnoucement printOn: self stream. > self stream cr > > to emphasise the double dispatch. Good catch! Committed. Cheers, Doru > Cheers, > > Doru > > > > -- > > www.tudorgirba.com > > > > "Every thing has its own flow" > > > -- www.tudorgirba.com "Every thing has its own flow"

June 16, 2014

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by Yuriy Tymchuk

Ah, no, itâs my fault. Sometimes I implicitly skip the middle of a sentence :). Uko On 16 Jun 2014, at 10:34, kilon alios <kilon.alios(a)gmail.com> wrote: > "I could not agree more with you" means that I agree with you 100%. Its one of the strange things of language to use a negative to imply a positive statement :D > > > On Mon, Jun 16, 2014 at 11:09 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > @Kilon, you said almost the same thing, how comes that you do not agree with me? > > Uko > > > On 16 Jun 2014, at 10:03, kilon alios <kilon.alios(a)gmail.com> wrote: > >> I could not agree more with you >> >> I am very much against "One ring to Rule them All" , it did not work for Sauron in Lord of the Rings for the exact reason why its a bad idea to put all your eggs in one basket. Once it fails in a fundamental level you are screwed forever. >> >> People reinvent the wheel because this is how evolution works, this is how our world works. You use what you have and you add something new to the recipe and hope for the best. >> >> It also gives the ability to users to choose the right tool for them. A developer of a Library can never be more smart than its users because he does not know their needs and cannot anticipate their personal desires and workflows. >> >> I have never had the pleasure of using a Logging system, but using Announcements sounds like a way I would go with it if it was me. Certainly will give it a try thanks Tudor. >> >> >> On Mon, Jun 16, 2014 at 10:13 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: >> >> On 16 Jun 2014, at 08:52, stepharo <stepharo(a)free.fr> wrote: >> >>> >>>> Hi, >>>> >>>> I like very much the new energy people are putting into creating the SystemLogger engine for Pharo. I think this is a specifically important area for which we have to have a solution out of the box. At the same time, I also think that Pharo provides an infrastructure that makes room for ideas that are otherwise hard to reach in other languages or environments. >>> >>> Why Java does not have announcements? >>> >>>> Stef asked for collaborations around this project, so here is my literally small contribution: a rather different logging engine. >>> I do not see how this contribute to SystemLogger. So at least please do not say it, respect the amount of time I spent >>> design it and working with Norbert. >>> >>>> It is called Beacon, it is based entirely on Announcements, it has ~200 lines of code, it has no tags or levels, and in my opinion it is fully functional. >>>> >>>> You can see a detailed description here including some informal comparisons with SystemLogger: >>>> http://www.humane-assessment.com/blog/beacon >>>> >>>> Please let me know what you think. I would be happy to join forces to reach a mature solution that is both versatile and that can show how Pharo is different. >>> So should we see it as a competitor to SystemLogger? >>> (you will say of course not) but I do not understand. >> >> Just my 2 cents: >> Stef, I wouldnât fight agains other projects, itâs natural that for popular environments there are different approaches to implement important tools. E.g. in Ruby there are few different web frameworks and I think logging too. And I wouldnât be angry on Doru because he mentioned SystemLogger, for me the idea is that Pharo is missing decent logger, and there is a work being done on the big project, in a while Doru proposes small yet functional tool with itâs own philosophy. >> >> Iâm trying to ignore my personal preferences here, and move attention to the point that we should make our environment modular and let people decide what they want to use. >> >> Uko >> >>> >>> >>>> >>>> Cheers, >>>> Doru >>>> >>>> -- >>>> www.tudorgirba.com >>>> >>>> "Every thing has its own flow" >>> >> >> > >

June 16, 2014

Re: [Pharo-dev] Where is the postcard with syntax?

by stepharo

you have the flyer of Damien (no idea where it is) but no real postcard. Stef On 16/6/14 09:35, Yuriy Tymchuk wrote: > Hi guys, > > we all are talking about the syntax fitting in a postcard, but was there any real postcard with Pharo syntax prototype? This would be really interesting. > > Uko > >

June 16, 2014

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by kilon alios

"I could not agree more with you" means that I agree with you 100%. Its one of the strange things of language to use a negative to imply a positive statement :D On Mon, Jun 16, 2014 at 11:09 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > @Kilon, you said almost the same thing, how comes that you do not agree > with me? > > Uko > > > On 16 Jun 2014, at 10:03, kilon alios <kilon.alios(a)gmail.com> wrote: > > I could not agree more with you > > I am very much against "One ring to Rule them All" , it did not work for > Sauron in Lord of the Rings for the exact reason why its a bad idea to put > all your eggs in one basket. Once it fails in a fundamental level you are > screwed forever. > > People reinvent the wheel because this is how evolution works, this is how > our world works. You use what you have and you add something new to the > recipe and hope for the best. > > It also gives the ability to users to choose the right tool for them. A > developer of a Library can never be more smart than its users because he > does not know their needs and cannot anticipate their personal desires and > workflows. > > I have never had the pleasure of using a Logging system, but using > Announcements sounds like a way I would go with it if it was me. Certainly > will give it a try thanks Tudor. > > > On Mon, Jun 16, 2014 at 10:13 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> > wrote: > >> >> On 16 Jun 2014, at 08:52, stepharo <stepharo(a)free.fr> wrote: >> >> >> Hi, >> >> I like very much the new energy people are putting into creating the >> SystemLogger engine for Pharo. I think this is a specifically important >> area for which we have to have a solution out of the box. At the same time, >> I also think that Pharo provides an infrastructure that makes room for >> ideas that are otherwise hard to reach in other languages or environments. >> >> >> Why Java does not have announcements? >> >> Stef asked for collaborations around this project, so here is my >> literally small contribution: a rather different logging engine. >> >> I do not see how this contribute to SystemLogger. So at least please do >> not say it, respect the amount of time I spent >> design it and working with Norbert. >> >> It is called Beacon, it is based entirely on Announcements, it has ~200 >> lines of code, it has no tags or levels, and in my opinion it is fully >> functional. >> >> You can see a detailed description here including some informal >> comparisons with SystemLogger: >> http://www.humane-assessment.com/blog/beacon >> >> Please let me know what you think. I would be happy to join forces to >> reach a mature solution that is both versatile and that can show how Pharo >> is different. >> >> So should we see it as a competitor to SystemLogger? >> (you will say of course not) but I do not understand. >> >> >> Just my 2 cents: >> Stef, I wouldnât fight agains other projects, itâs natural that for >> popular environments there are different approaches to implement important >> tools. E.g. in Ruby there are few different web frameworks and I think >> logging too. And I wouldnât be angry on Doru because he mentioned >> SystemLogger, for me the idea is that Pharo is missing decent logger, and >> there is a work being done on the big project, in a while Doru proposes >> small yet functional tool with itâs own philosophy. >> >> Iâm trying to ignore my personal preferences here, and move attention to >> the point that we should make our environment modular and let people decide >> what they want to use. >> >> Uko >> >> >> >> >> Cheers, >> Doru >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> >> >> >> > >

June 16, 2014

Re: [Pharo-dev] Performance for Video Games(Woden Video)

by kilon alios

I think its awesome that you try to make a 3d game engine for Pharo. I know its not an easy task by a long mark, its tons of hard work to have something reliable. Thank you for your efforts and if you want a beta tester I will be more than happy to help you. Saying that I am highly skeptical . There is a reason why 3d engines rarely become popular in dynamic languages. Performance. Same reason why C/C++dominate 3d graphics and audio synthesis. I agree that sending things to GPU will greatly boost performance for Pharo not just for Graphics but for all sort of things. CUDA and OpenCL can be used for loads of different computations and a good modern GPU can offer 8 times the performance of a good CPU. On the other hand having a good 3d game engine for Pharo is a thousand times better than not having such a thing at all. So keep rocking :) On Fri, Jun 13, 2014 at 6:16 AM, Ronie Salgado <roniesalg(a)gmail.com> wrote: > I am working on making a video game engine in Pharo. Recently added a FPS > counter and recorded some samples: http://youtu.be/-2ida5Q1mbg . > > Lets say I was quite surprised, so I have yet a lot more to optimize, but > a good trick seems to be making some computation in the GPU. > > In this video, I show 6 early examples. The water example is still > incomplete, missing reflection/refractions, but it shows what happens when > I compute the water waves using a shader in the GPU. > > Please, pay a close attention to the FPS count. > > BTW, it feels slow because of the screen capture and the tearing is there > because I haven't enabled vsync, but I have support for it. > > Greetings, > Ronie Salgado >

June 16, 2014

Re: [Pharo-dev] Vm on Mac OS 7.5 no cairo

by Sven Van Caekenberghe

On 16 Jun 2014, at 10:01, Stephan Eggermont <stephan(a)stack.nl> wrote: > I find I'm unable to use a recent vm with Roassal/cairo on Mac OS 7.5. I am assuming this is a typo, right ? Mac OS System 7.5 is from September 1994 ... http://en.wikipedia.org/wiki/System_7#Version_history > The dec 12 version works, but all new ones say: failed to get a symbol address: cairo_image_surface_create > > Stephan

June 16, 2014

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by Yuriy Tymchuk

@Kilon, you said almost the same thing, how comes that you do not agree with me? Uko On 16 Jun 2014, at 10:03, kilon alios <kilon.alios(a)gmail.com> wrote: > I could not agree more with you > > I am very much against "One ring to Rule them All" , it did not work for Sauron in Lord of the Rings for the exact reason why its a bad idea to put all your eggs in one basket. Once it fails in a fundamental level you are screwed forever. > > People reinvent the wheel because this is how evolution works, this is how our world works. You use what you have and you add something new to the recipe and hope for the best. > > It also gives the ability to users to choose the right tool for them. A developer of a Library can never be more smart than its users because he does not know their needs and cannot anticipate their personal desires and workflows. > > I have never had the pleasure of using a Logging system, but using Announcements sounds like a way I would go with it if it was me. Certainly will give it a try thanks Tudor. > > > On Mon, Jun 16, 2014 at 10:13 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > > On 16 Jun 2014, at 08:52, stepharo <stepharo(a)free.fr> wrote: > >> >>> Hi, >>> >>> I like very much the new energy people are putting into creating the SystemLogger engine for Pharo. I think this is a specifically important area for which we have to have a solution out of the box. At the same time, I also think that Pharo provides an infrastructure that makes room for ideas that are otherwise hard to reach in other languages or environments. >> >> Why Java does not have announcements? >> >>> Stef asked for collaborations around this project, so here is my literally small contribution: a rather different logging engine. >> I do not see how this contribute to SystemLogger. So at least please do not say it, respect the amount of time I spent >> design it and working with Norbert. >> >>> It is called Beacon, it is based entirely on Announcements, it has ~200 lines of code, it has no tags or levels, and in my opinion it is fully functional. >>> >>> You can see a detailed description here including some informal comparisons with SystemLogger: >>> http://www.humane-assessment.com/blog/beacon >>> >>> Please let me know what you think. I would be happy to join forces to reach a mature solution that is both versatile and that can show how Pharo is different. >> So should we see it as a competitor to SystemLogger? >> (you will say of course not) but I do not understand. > > Just my 2 cents: > Stef, I wouldnât fight agains other projects, itâs natural that for popular environments there are different approaches to implement important tools. E.g. in Ruby there are few different web frameworks and I think logging too. And I wouldnât be angry on Doru because he mentioned SystemLogger, for me the idea is that Pharo is missing decent logger, and there is a work being done on the big project, in a while Doru proposes small yet functional tool with itâs own philosophy. > > Iâm trying to ignore my personal preferences here, and move attention to the point that we should make our environment modular and let people decide what they want to use. > > Uko > >> >> >>> >>> Cheers, >>> Doru >>> >>> -- >>> www.tudorgirba.com >>> >>> "Every thing has its own flow" >> > >

June 16, 2014

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by kilon alios

I could not agree more with you I am very much against "One ring to Rule them All" , it did not work for Sauron in Lord of the Rings for the exact reason why its a bad idea to put all your eggs in one basket. Once it fails in a fundamental level you are screwed forever. People reinvent the wheel because this is how evolution works, this is how our world works. You use what you have and you add something new to the recipe and hope for the best. It also gives the ability to users to choose the right tool for them. A developer of a Library can never be more smart than its users because he does not know their needs and cannot anticipate their personal desires and workflows. I have never had the pleasure of using a Logging system, but using Announcements sounds like a way I would go with it if it was me. Certainly will give it a try thanks Tudor. On Mon, Jun 16, 2014 at 10:13 AM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: > > On 16 Jun 2014, at 08:52, stepharo <stepharo(a)free.fr> wrote: > > > Hi, > > I like very much the new energy people are putting into creating the > SystemLogger engine for Pharo. I think this is a specifically important > area for which we have to have a solution out of the box. At the same time, > I also think that Pharo provides an infrastructure that makes room for > ideas that are otherwise hard to reach in other languages or environments. > > > Why Java does not have announcements? > > Stef asked for collaborations around this project, so here is my > literally small contribution: a rather different logging engine. > > I do not see how this contribute to SystemLogger. So at least please do > not say it, respect the amount of time I spent > design it and working with Norbert. > > It is called Beacon, it is based entirely on Announcements, it has ~200 > lines of code, it has no tags or levels, and in my opinion it is fully > functional. > > You can see a detailed description here including some informal > comparisons with SystemLogger: > http://www.humane-assessment.com/blog/beacon > > Please let me know what you think. I would be happy to join forces to > reach a mature solution that is both versatile and that can show how Pharo > is different. > > So should we see it as a competitor to SystemLogger? > (you will say of course not) but I do not understand. > > > Just my 2 cents: > Stef, I wouldnât fight agains other projects, itâs natural that for > popular environments there are different approaches to implement important > tools. E.g. in Ruby there are few different web frameworks and I think > logging too. And I wouldnât be angry on Doru because he mentioned > SystemLogger, for me the idea is that Pharo is missing decent logger, and > there is a work being done on the big project, in a while Doru proposes > small yet functional tool with itâs own philosophy. > > Iâm trying to ignore my personal preferences here, and move attention to > the point that we should make our environment modular and let people decide > what they want to use. > > Uko > > > > > Cheers, > Doru > > -- > www.tudorgirba.com > > "Every thing has its own flow" > > > >

June 16, 2014

Vm on Mac OS 7.5 no cairo

by Stephan Eggermont

I find I'm unable to use a recent vm with Roassal/cairo on Mac OS 7.5. The dec 12 version works, but all new ones say: failed to get a symbol address: cairo_image_surface_create Stephan

June 16, 2014

Re: [Pharo-dev] [ann] beacon - a slim announcement-based logging engine

by Sven Van Caekenberghe

On 15 Jun 2014, at 15:04, Tudor Girba <tudor(a)tudorgirba.com> wrote: > Hi, > > I like very much the new energy people are putting into creating the SystemLogger engine for Pharo. I think this is a specifically important area for which we have to have a solution out of the box. At the same time, I also think that Pharo provides an infrastructure that makes room for ideas that are otherwise hard to reach in other languages or environments. > > Stef asked for collaborations around this project, so here is my literally small contribution: a rather different logging engine. It is called Beacon, it is based entirely on Announcements, it has ~200 lines of code, it has no tags or levels, and in my opinion it is fully functional. > > You can see a detailed description here including some informal comparisons with SystemLogger: > http://www.humane-assessment.com/blog/beacon > > Please let me know what you think. I would be happy to join forces to reach a mature solution that is both versatile and that can show how Pharo is different. Here is some of my feedback: - Like I said in my document and like I tried to show in my implementation for Zinc, I want framework agnostic pure object logging, so I don't want to subclass from anything (I do from Announcement, but that is an implementation detail, the technique how #emit is done), can you deal with the ZnNewLogEvents as the are ? - I find what you did in StackSignal very interesting, when putting Exceptions in log objects, a natural thing to do, the same problems occur. I think we are just exploring this area and have lots to learn (debugging in the past). - The fact that FuelBeacon creates a new file for every signal does not strike me as efficient enough for real world use. Also, using the #printString as filename is weird for me. - I am also wondering whether the fact that Announcements are synchronous (I mean distributed to consumers on the same thread as the one that produced them) is good or not because like this, slow backend processing for whatever reason will kill performance of the producing code. - In your blog post (very good as usual ;-) I am still bothered by the usage of strings in the examples (I know you also use the stack signal, but that is more advanced). That way, reader will still think about text. We need better examples that show the power of object logging. I'll see if I can come up with something. - Small code change StringStreamBeacon>>#handle: anAnnoucement self stream nextPutAll: anAnnoucement printString; cr would be more efficient (no intermediate string) as StringStreamBeacon>>#handle: anAnnoucement self stream print: anAnnoucement; cr or even StringStreamBeacon>>#handle: anAnnoucement anAnnoucement printOn: self stream. self stream cr to emphasise the double dispatch. > Cheers, > Doru > > -- > www.tudorgirba.com > > "Every thing has its own flow"

June 16, 2014