- Pharo-dev - lists.pharo.org

[Pharo-project] Morphic 3, Cuis, Simple Morphic and Esug's sponsoring
by Juan Vuletich Nov. 7, 2010

Nov. 7, 2010

Hi Folks, All what Stef said wrt these issues is correct. I just want to (perhaps) make some detail clearer. Esug agreed to sponsor Morphic 3 about one year ago. At that time I expected to have a working and usable system by the end of this year. This is not going to happen. Main reasons are: 1) I underestimated the effort. 2) Inflation and other issues meant I needed to make more money that I was currently making. So I had to reduce the hours I devote to volunteer work to spend them in paid work. 3) I failed to get sponsors for this project outside the Smalltalk community (although I did try). 4) Some of the ideas in Morphic 3 are novel enough that I need to protect them from being patented by others from, before disclosing them. I don't like software patents, so I believe the best way to do this (and get more people to know these ideas) is to submit a paper to a journal such as ACM Transactions on Graphics. This will obviously require additional time. So work on Morphic 3 is not stopped, but right now it is paused. It seemed to me that instead of just telling Esug "I'm sorry, but you'll get nothing by the date I promised results", I could do a more practical, down to earth contribution to the Smalltalk community in return for Esug's support. I believe that many in the Pharo community share my opinion that the current Morphic implementation is obfuscated, that it is too hard to write applications with it, and that it is too hard to fix bugs and evolve it. The Morphic implementation in Cuis is the main result of many (I guess about 6) years of part time work on simplifying the system. I think it could be of great value for Pharo, and an enabler for evolution. So, I did a request to Esug, to deliver this, naming it "SimpleMorphic", ready to run in Pharo, instead of Morphic 3. Esug approved it, and I thank them for this. SimpleMorphic is the result of this. I started by extracting the Morphic in Cuis. This was not trivial, as I had to think carefully where to draw the line between what I'd migrate to Pharo and what I'd use from Pharo. Then I adapted it to some differences between Cuis and Pharo, and also tweaked Pharo to enable using alternative UIs. I worked about 80 hours on this and delivered this week. After all this, I think it is fair to say that Esug sponsored Cuis as SimpleMorphic is part of it. I hope this helps and does not add confusion. Cheers, Juan Vuletich

1 0

[Pharo-project] SmalltalkImage
by Benjamin Nov. 7, 2010

Nov. 7, 2010

Hi guys, I wonder how SmalltalkImage create the Smalltalk instance. Can someone explain me that ? Thank you Ben

2 1

[Pharo-project] OSX Pharo Shout Workspace is laggy
by sergio_101 Nov. 7, 2010

Nov. 7, 2010

i just downloaded the latest pharo image, and everything seems good, but the problem i am having is that when the dropdowns appear for tab completion, there is a noticeable lag in the system. it takes a few seconds to select one of the drop down items.. about four seconds or so.. my machine is a newer high powered macbook pro.. has anyone seen anything like this before? thanks! -- ---- peace, sergio photographer, journalist, visionary http://www.CodingForHire.com http://www.coffee-black.com http://www.painlessfrugality.com http://www.twitter.com/sergio_101 http://www.facebook.com/sergio101

2 4

[Pharo-project] The next milestone reached!
by Pavel Krivanek Nov. 6, 2010

Nov. 6, 2010

Hi all, I successfully loaded the the rest of PharoCore packages into the PharoKernel. The Morphic initialization works too. See the attached screenshot. I will publish the scripts later today. Cheers, -- Pavel

7 8

[Pharo-project] RoarVM - Pharo and Squeak on Multicore
by Torsten Bergmann Nov. 6, 2010

Nov. 6, 2010

Read more: http://astares.blogspot.com/2010/11/roarvm-pharo-and-squeak-on-multicore.ht… -- GRATIS! Movie-FLAT mit Ã¼ber 300 Videos. Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome

8 11

Re: [Pharo-project] About simple morphic
by StÃ©phane Ducasse Nov. 6, 2010

Nov. 6, 2010

On Nov 6, 2010, at 7:56 PM, Tudor Girba wrote: > Hi, > > I would be interested in testing these changes, but could we get a summary, or at least the intention behind these changes? What does Simple mean in this context? less dust in morphic. This is basically the work that juan did in CUIS now I do not have a detailled list of the changes. > > Cheers, > Doru > > > On 6 Nov 2010, at 19:10, StÃ©phane Ducasse wrote: > >> Thanks german >> This is what I think too but I like that other people express what they think :) >> So thanks. >> >> Stef >> >> On Nov 6, 2010, at 7:00 PM, GermÃ¡n Arduino wrote: >> >>> I'm not sure if my opinion can have some value because I'm more on the >>> web arena than in en desktop, but I've on my plans make somethings >>> (when time permit) using Morphic (Most rewriting things I did in the >>> past with Squeak). >>> >>> I think that if SimpleMorphic is loadable/unloadable and don't >>> generates bad effects in the Pharo environment, the proposal sound >>> correct and is a good way to have a Morphic simplification (A thing I >>> think we need). >>> >>> Just my opinion. >>> GermÃ¡n. >>> >>> >>> 2010/11/6 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >>>> Hi guys >>>> >>>> the mail may be a bit long sorry. ESUG was supporting Morphic 3 (a new proposal) of Juan. Now it occurred that Juan realized that he will not be able to deliver what he wanted. So After discussion one idea was to get simpleMorphic (not lighMorphic) a simpler version of Morphic done by juan for Squeak and Pharo. We believe that this is good because Juan could deliver something and we could support him. >>>> >>>> Juan sent us the code and I published it at http://code.google.com/p/pharo/issues/detail?id=3210 >>>> >>>> Here is the original text: >>>> >>>>> I'm really happy to tell you: SimpleMorphic for Pharo is working! The attached zip file contains a few numbered change sets, to be installed in Pharo in order. >>>>> >>>>> The contents of the change sets are as follows: >>>>> - SMx01-PharoEnhAndPreliminaries-jmv.3.cs and SMx02-PharoEnhAndPreliminariesCoda.1.cs contain some refactoring needed to make it possible to add another GUI to Pharo. Nothing here is specific to SimpleMorphic, and I recommend integrating them into Pharo. >>>>> - SMx03-ExtensionMethodsPart1.3.cs contains a few useful methods, not specific to SimpleMorphic, but optional anyway. You might integrate this into Pharo or consider it part of SimpleMorphic. >>>>> - SMx04-SimpleMorphic.11.cs This change set is the big one. It includes all the new SimpleMorphic classes. It doesn't include any method for existing classes (to make review easier). >>>>> - SMx05-ExtensionMethodsPart2.7.cs This change set includes the SimpleMorphic specific methods I added to existing classes. >>>>> >>>>> After loading them, you can evaluate 'SimpleMorphicUIManager start' to enter SimpleMorphic and 'SimpleMorphicUIManager stop' to return to regular Morphic. >>>>> >>>>> As SimpleMorphic is much simpler and easier to understand that regular Morphic, my biggest hope is that the Pharo community will adopt it, study it in detail, and be able to adapt it to Pharo's needs all way down. No need to walk away from overwhelming complexity anymore. >>>>> >>>>> I look forward for your comments! >>>> >>>> >>>> Now after discussion with Juan: we plan to integrate SM01, 02, and 03 to be able to load SM04. >>>> SM04 is not in a shape were we can simply migrate on it >>>> - Polymorph first polymorph is tight with the old Morphic >>>> - Preference are still used >>>> - FillInTheBlank, PopUp.... are used. >>>> >>>> Now I suggest that we integrate 01.02.03 and package the rest to be loaded in 1.2 >>>> This way we can either update SimpleMorphic to get more pharoed or we could use it to study and fix morphic. >>>> But all that will require some work. >>>> >>>> Tell us what you think. >>>> >>> >> >> > > -- > www.tudorgirba.com > > "Some battles are better lost than fought." > > > >

2 1

Re: [Pharo-project] The next milestone reached!
by StÃ©phane Ducasse Nov. 6, 2010

Nov. 6, 2010

Gabriel, Benjamin and Nicolas paez started to work on kernels (collaborating also with pavel). Now Benjamin is starting to generate a new kernel using microSqueak approach. just that you know our plans we are working on pine kernel = microsqueak with as less as modifications as possible (fun for teaching) ---------------------------------- hazelnut = nothing to do with pine kernel but inspired and based on current pharo kernel String wideString floats... Exception walnut = a bit bigger but based on haselnuts Announcement .... coconut = probably bigger but based on I imagine you got the idea of our metaphor :) In parallel we are cleaning cleaning cleaning and repackaging part of the system in collaboration with Pavel so that when we get haselnuts or walnuts we could do something with it. and we have some other cool stuff on the grill :) Stef

1 0

Nov. 6, 2010

Thanks german This is what I think too but I like that other people express what they think :) So thanks. Stef On Nov 6, 2010, at 7:00 PM, GermÃ¡n Arduino wrote: > I'm not sure if my opinion can have some value because I'm more on the > web arena than in en desktop, but I've on my plans make somethings > (when time permit) using Morphic (Most rewriting things I did in the > past with Squeak). > > I think that if SimpleMorphic is loadable/unloadable and don't > generates bad effects in the Pharo environment, the proposal sound > correct and is a good way to have a Morphic simplification (A thing I > think we need). > > Just my opinion. > GermÃ¡n. > > > 2010/11/6 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> Hi guys >> >> the mail may be a bit long sorry. ESUG was supporting Morphic 3 (a new proposal) of Juan. Now it occurred that Juan realized that he will not be able to deliver what he wanted. So After discussion one idea was to get simpleMorphic (not lighMorphic) a simpler version of Morphic done by juan for Squeak and Pharo. We believe that this is good because Juan could deliver something and we could support him. >> >> Juan sent us the code and I published it at http://code.google.com/p/pharo/issues/detail?id=3210 >> >> Here is the original text: >> >>> I'm really happy to tell you: SimpleMorphic for Pharo is working! The attached zip file contains a few numbered change sets, to be installed in Pharo in order. >>> >>> The contents of the change sets are as follows: >>> - SMx01-PharoEnhAndPreliminaries-jmv.3.cs and SMx02-PharoEnhAndPreliminariesCoda.1.cs contain some refactoring needed to make it possible to add another GUI to Pharo. Nothing here is specific to SimpleMorphic, and I recommend integrating them into Pharo. >>> - SMx03-ExtensionMethodsPart1.3.cs contains a few useful methods, not specific to SimpleMorphic, but optional anyway. You might integrate this into Pharo or consider it part of SimpleMorphic. >>> - SMx04-SimpleMorphic.11.cs This change set is the big one. It includes all the new SimpleMorphic classes. It doesn't include any method for existing classes (to make review easier). >>> - SMx05-ExtensionMethodsPart2.7.cs This change set includes the SimpleMorphic specific methods I added to existing classes. >>> >>> After loading them, you can evaluate 'SimpleMorphicUIManager start' to enter SimpleMorphic and 'SimpleMorphicUIManager stop' to return to regular Morphic. >>> >>> As SimpleMorphic is much simpler and easier to understand that regular Morphic, my biggest hope is that the Pharo community will adopt it, study it in detail, and be able to adapt it to Pharo's needs all way down. No need to walk away from overwhelming complexity anymore. >>> >>> I look forward for your comments! >> >> >> Now after discussion with Juan: we plan to integrate SM01, 02, and 03 to be able to load SM04. >> SM04 is not in a shape were we can simply migrate on it >> - Polymorph first polymorph is tight with the old Morphic >> - Preference are still used >> - FillInTheBlank, PopUp.... are used. >> >> Now I suggest that we integrate 01.02.03 and package the rest to be loaded in 1.2 >> This way we can either update SimpleMorphic to get more pharoed or we could use it to study and fix morphic. >> But all that will require some work. >> >> Tell us what you think. >> >

3 2

Nov. 6, 2010

> A great job is done in aggregating energy to make Pharo a sexy Smalltalk > development plateform. > Was SimpleMorphic sponsoried under this perspective? not really. ESUG wanted to push Morphic 30 and juan did a new proposal (one that was more reasonable than the one in 2008). Now since having a cleaner morphic is important we found it a good fallback (making sure that the work done in cuis could be loaded in other systems so that they can build on it or get influenced). Now we have to have a look and decide what to do. > And more importantly what is the outcome for a Desktop application developer like me? For now not something directly usable if you want to use polymorph for example. as I said this is our turn to make some use of it. > > Hilaire > > Le 06/11/2010 19:10, StÃ©phane Ducasse a Ã©crit : >> Thanks german >> This is what I think too but I like that other people express what they think :) >> So thanks. >> >> Stef >> >> On Nov 6, 2010, at 7:00 PM, GermÃ¡n Arduino wrote: >> >>> I'm not sure if my opinion can have some value because I'm more on the >>> web arena than in en desktop, but I've on my plans make somethings >>> (when time permit) using Morphic (Most rewriting things I did in the >>> past with Squeak). >>> >>> I think that if SimpleMorphic is loadable/unloadable and don't >>> generates bad effects in the Pharo environment, the proposal sound >>> correct and is a good way to have a Morphic simplification (A thing I >>> think we need). >>> >>> Just my opinion. >>> GermÃ¡n. >>> >>> >>> 2010/11/6 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >>>> Hi guys >>>> >>>> the mail may be a bit long sorry. ESUG was supporting Morphic 3 (a new proposal) of Juan. Now it occurred that Juan realized that he will not be able to deliver what he wanted. So After discussion one idea was to get simpleMorphic (not lighMorphic) a simpler version of Morphic done by juan for Squeak and Pharo. We believe that this is good because Juan could deliver something and we could support him. >>>> >>>> Juan sent us the code and I published it at http://code.google.com/p/pharo/issues/detail?id=3210 >>>> >>>> Here is the original text: >>>> >>>>> I'm really happy to tell you: SimpleMorphic for Pharo is working! The attached zip file contains a few numbered change sets, to be installed in Pharo in order. >>>>> >>>>> The contents of the change sets are as follows: >>>>> - SMx01-PharoEnhAndPreliminaries-jmv.3.cs and SMx02-PharoEnhAndPreliminariesCoda.1.cs contain some refactoring needed to make it possible to add another GUI to Pharo. Nothing here is specific to SimpleMorphic, and I recommend integrating them into Pharo. >>>>> - SMx03-ExtensionMethodsPart1.3.cs contains a few useful methods, not specific to SimpleMorphic, but optional anyway. You might integrate this into Pharo or consider it part of SimpleMorphic. >>>>> - SMx04-SimpleMorphic.11.cs This change set is the big one. It includes all the new SimpleMorphic classes. It doesn't include any method for existing classes (to make review easier). >>>>> - SMx05-ExtensionMethodsPart2.7.cs This change set includes the SimpleMorphic specific methods I added to existing classes. >>>>> >>>>> After loading them, you can evaluate 'SimpleMorphicUIManager start' to enter SimpleMorphic and 'SimpleMorphicUIManager stop' to return to regular Morphic. >>>>> >>>>> As SimpleMorphic is much simpler and easier to understand that regular Morphic, my biggest hope is that the Pharo community will adopt it, study it in detail, and be able to adapt it to Pharo's needs all way down. No need to walk away from overwhelming complexity anymore. >>>>> >>>>> I look forward for your comments! >>>> >>>> >>>> Now after discussion with Juan: we plan to integrate SM01, 02, and 03 to be able to load SM04. >>>> SM04 is not in a shape were we can simply migrate on it >>>> - Polymorph first polymorph is tight with the old Morphic >>>> - Preference are still used >>>> - FillInTheBlank, PopUp.... are used. >>>> >>>> Now I suggest that we integrate 01.02.03 and package the rest to be loaded in 1.2 >>>> This way we can either update SimpleMorphic to get more pharoed or we could use it to study and fix morphic. >>>> But all that will require some work. >>>> >>>> Tell us what you think. >>>> >>> >> >> >> > > > -- > Dr. Geo, to discover geometry on Linux, Windows, MAC and XO > http://community.ofset.org/index.php/DrGeo > >

1 0

Nov. 6, 2010

On Nov 6, 2010, at 5:55 PM, Levente Uzonyi wrote: > On Sat, 6 Nov 2010, StÃ©phane Ducasse wrote: > >>> you can add ESUG as sponsors >>> >>> Is the developement of Cuis sponsored by ESUG? >> >> We originally sponsored the development of Morphic 30 and it generally does not hurt to say it. >> Even if at the end morphic is not ready for public consumption. Our original idea was not to >> sponsor SimpleMorphic. I have no problem with the solution we found because he helps everybody to be happy. > > Did Juan stop developing Morphic 3? I doubt he did. As I told you juan decided that Morphic 30 would not be ready and he preferred to find another solution for the sponsor we got with him. Any board member can confirm that and juan too. We found that the solution to get a simplemorphic was a good deal so that Juan does not feel embarassed about esug and the inverse. Now you that believe what you want but I do not believe that this is really constructive. Stef > > > Levente > >> >> Stef > > snip

2 1